Your potential is your power.

I’ve reread it more times than I can count. Mostly on the days I doubted myself.

It took me a while to see the part I kept skipping. Potential isn’t power on its own. It becomes power when you’re aware of it. And when you act on it.

The story we tell ourselves

Everyone wants the same things. Reach our goals. Be financially free. Be fit in body and mind. Have relationships that work. Live a life that feels full.

Most people believe that kind of life happens to a selected few. The talented ones. The lucky ones.

It doesn’t.

Read the stories of people who built something great. So many of them started with literally nothing. No money. No obvious talent for what they were doing. Somewhere along the way, something changed.

Not their situation. Their thinking.

The quiet trap

Most of us live in the middle. Not failing. Not flying.

The cafe stays an idea. The business stays a someday.

We tell ourselves we’ll start when we’re ready. But if you don’t believe in yourself, why would anyone else?

The loud trap

Then there’s the opposite problem. Plenty of energy, pointed everywhere.

One thing doesn’t work out, so we hop to the next. A new idea. A new course. A new niche.

Sunlight is everywhere, and it can’t burn a sheet of paper. Hold a magnifying glass under it and it will. A laser cuts through steel for the same reason: everything on one point.

Energy isn’t the problem. Focus is.

One question worth asking

What is my strength, and am I putting my time there?

You become the best at what you do when you work at the edge of your strength. Not when you spread yourself across ten things you’re average at.

Choose your focus. Choose your niche. Then give it the time it deserves.

Would you drive that car?

Imagine you don’t trust your car’s brakes. Would you drive it? Maybe. Slowly. Holding back the whole way.

That’s what doubt does to a dream. You never stop, but you never really go either.

Believing in your potential isn’t wishful thinking. It’s what lets you take the action that brings the results. Even when they show up later than you’d like.

The line I’m leaving you with

Your only limits are the ones you set yourself.

Believe in your potential. Back your strength. And don’t let self-doubt hold you captive.

Whatever matters to you, start there.

Rayn

Listen to the full episode, Ep.56: Your potential is your power, on Whatever Matters with Rayn.

be very ambitious in the head, but very realistic in the daily practice.

Simple suggestion, hard when you start to executive or live it.

1. Everyone’s ambitious. That part’s easy.

Wanting a big outcome might not cost anyone who’s reading this anything. If you write a book, the company or a startup that you build, if you want to run a marathon, or I don’t know, whatever your goals are. Everyone wants something big, something ambitious, something beyond their reach at the moment, right? It doesn’t set you apart from the next person with the same goal because they also have similar ambitions, and that part is the hardest for people to remember.

2. The daily practice is where people trip up.

I watched this happen to friends and some close founder folks. Sometimes I’ve done this to myself. I set a huge goal. I sit down and plan, if I have that time, for a day, for a week, maybe for a month, and I build a plan that only works for someone who’s achieved it, right?

2,000 words of writing every day would mean I’m writing 10,000 words a week, and that is action orientation. It’s really not a plan, so I think a lot of us confuse what planning means. Planning is to set habits. Planning is to be disciplined. Planning is to have action items instead of the execution itself.

3. There’s supposed to be a gap.

For a long time I thought the distance between where I wanted to be and what I was doing today was something to close immediately. It isn’t. James Clear put this better than I can in Atomic Habits: “You do not rise to the level of your goals. You fall to the level of your systems.” Your daily practice is the system. Build it for who you actually are today, not who you’re trying to become.

4. Make the practice something you can actually keep.
As I gave the example of writing a few thousand words every day, because our goals are ambitious and the end motive is to achieve big things, we end up forgetting that the broken-down action items or the broken-down tasks that we do have to be very small. Instead of saying 10,000 a week or 2,000 a day, you start with 15 minutes of writing. The number itself is smaller. Your brain can comprehend that this is achievable, hence action is easier.

This can correlate very quickly with how long a short-form video is today. It could be 30 seconds, could be 90 seconds. It is a very small amount of time from your 24 hours a day, but you end up doing that so much, so often, because the minute task of swiping up is probably just a millisecond. That is where practice comes in handy. Our brain gets trained when the broken-down tasks are way smaller, way minute, and they could just come together over a period of 3 months. If you are smart enough, you will be able to figure that out easily.

5. Keep the dream loud and the daily habit boring.

That’s not a contradiction, it’s the whole approach. Big vision up top, small repeatable steps underneath, and enough patience to let the second one eventually catch up with the first.

I’m rebuilding a few of my own habits around this right now. Loud ambition, boring practice. That’s probably the whole way to live this life.

We got onto the subject of confidence, and I asked him straight up where it actually comes from.

His answer stuck with me because it wasn’t what I expected. He said confidence always comes from yesterday. If you’ve done things yesterday, you’ll be confident today. But the further away you get from the last thing you actually did, the less confident you feel. So that window, between your last action and today, has to stay very, very small.

That’s not a motivational line. That’s just how it works.

1. Stop treating confidence like a savings account.

I used to think if I banked enough wins early on, I’d be set for a while. Point back to something I did months ago as proof I’m still capable. It doesn’t work like that. The marathon you ran two years ago won’t carry you through this week. The five kilometres you ran this morning will.

2. That’s a problem if you’ve been coasting.

If the last thing you’re proud of happened a while back, your confidence is running on fumes, even if it doesn’t feel that way yet.

3. It’s good news if you’re just starting out.

You don’t need a big, dramatic win to feel capable. You need tomorrow’s small thing, then the day after’s. That’s the whole mechanism.

4. Ask a different question.

Instead of “what have I achieved,” try “what did I do yesterday.” If that’s a hard one to answer honestly, you already know where the gap is.

Aristotle put it more elegantly than any podcast guest ever will: “We are what we repeatedly do. Excellence, then, is not an act but a habit.” Confidence works the same way. It’s not a trophy you keep on a shelf. It’s a habit you either keep up or let lapse.

I’m keeping my window small this week. What did you do yesterday?

I want to tell you about the exact moment I realised hitting a goal doesn’t feel the way you think it will. It happened to me a few years back, and it’s stuck with me since.

The day when I finished a 10K run under 60 minutes. That’s, I guess, the top 1% of the crowd or maybe top 3% of all the runners around the world.

I expected a wave of something. Relief, maybe. Pride. Instead I refreshed and went back about my day.

That’s the arrival fallacy.

A term coined by Tal Ben-Shahar, a positive psychology researcher who taught at Harvard, to describe the mistaken belief that reaching a goal will bring lasting happiness.

It doesn’t. Not because the goal was wrong. Because happiness was never sitting at the finish line waiting for you.

Nobody warns you about this part before you start chasing something big, something beyond you.

The finish line moves the second you touch it.

Get your first customer, you want a hundred. Reach $5000 in savings, you want $15000 next, close a seed round investment, you want to get to series A faster.

and from my founder experience. We got into Y Combinator, we wanted Demo Day to go well. Then close our seed raise. Then the next hire. Then the acquisition offer. So on and so forth.

I’ve watched this pattern in founder after founder, myself included. We treat every milestone like a summit. It isn’t one. It’s a base camp.

The Base Camp Thinking

Climbers don’t sprint from base camp to summit in one push. They stop. They acclimatise. They let their body catch up to where they are before pushing higher. They even climb down, and then climb back up. It’s only natural.

But most of us skip the acclimatisation step completely. We touch a goal and immediately start scanning for the next one, without ever letting ourselves feel like we arrived.

I recommend you try this instead the next time you’ve achieved a milestone. When you hit something you’ve been working toward, stop for a full day. Not a mental note to “enjoy it later.” An actual full day. Tell someone about it. Write it down in your journal. Let it be real before you let it become old news, before it becomes stale.

You’re allowed to decide a moment is enough, even when your ambition insists it isn’t.

I didn’t do that the day I finished my 10K run under 60 mins. I regret it more than almost any other decision that year. So I made a rule for myself: no new goal gets set until the last one has been sat with for at least 24 hours.

It’s a small rule but effective. It’s changed how the wins actually feel, it sits with you longer than you think.

Until next time, sit with your wins a little longer than feels comfortable.

— Rayn

Connect with me on LinkedIn

The founders who actually make it rarely get one.

When I built Unschool, we never had a single viral spike. No moment where the numbers exploded and stayed there.

We crossed 250,000 users and $1 million in revenue almost entirely through organic growth. One cohort at a time. One referral at a time.

For a long while, I thought that meant we were doing something wrong. Everyone else seemed to have a launch story. We just had a slow, steady climb that looked boring on a chart.

It wasn’t boring. It was the only thing that actually compounds.

A viral spike behaves like a firework. It shoots up fast, looks spectacular, and comes down just as fast. Most people who show up during a spike never come back. They saw something interesting once and moved on.

The spike changes your metrics for a week. It rarely changes your business.

Paul Graham wrote about this in 2013, in an essay called “Do Things that Don’t Scale.” His argument was blunt: almost every successful startup he’d funded at Y Combinator grew by doing something manual, unscalable, and unglamorous in its first year.

Airbnb’s founders went door to door in New York, photographing listings themselves. Stripe’s founders sat next to early users and installed the product on their laptops, on the spot.

None of that scales. All of it compounds.

That’s the shift worth making. Stop asking what your next big launch should be. Start asking what one motion you can repeat every single week, forever, that brings in the right ten customers.

That’s the difference between a spike and a slope. A spike is a moment you create once. A slope is a motion you repeat until it becomes your identity in the market.

I call this the weekly proof test. Pick one growth motion, cold outreach, a piece of content, a demo call, a partnership conversation, and run it every week without exception. Run it in the weeks it feels good. Run it in the weeks it feels like nothing is working.

Twelve weeks in, you’ll have a real signal. Twelve days in, you’ll have almost nothing, and that’s fine.

Founders who chase the spike quit around week three because the numbers look flat. Founders who commit to the slope are still standing at month twelve, with a business that doesn’t depend on luck.

You don’t need one big moment. You need one small motion, repeated until it compounds. Pick the one thing you can do every week for the next three months.

Then do it, especially in the weeks it feels pointless.

Until next time, keep showing up on the boring weeks too.

— Rayn Connect on LinkedIn → https://www.linkedin.com/in/chalkmeout

If you’re building the slope and want a second pair of eyes on it → https://topmate.io/narayanan

This week I want to talk about the job market in the AI era. Specifically: why most people’s current strategy is actively making things worse, and what the ones actually landing roles are doing instead.

TLDR

• 75% of CVs never reach a human recruiter, per Fortune’s March 2026 investigation

• Application volume surged 93% in twelve months while the offer rate across all applications sits at just 0.5%

• Salesforce CEO Marc Benioff confirmed in May that Salesforce is keeping engineering lean with AI, but actively hiring in sales

• Job postings requiring AI skills grew 144% year-on-year as of April 2026

• Net global job creation from AI disruption is projected at 78 million roles by 2030, but 92 million jobs will be displaced before that

A machine is deciding your future before a human has read your name

The Fortune headline in March said it plainly: 75% of CVs never reach a human recruiter. The ones that don’t make it aren’t rejected. They never enter the conversation at all.

Here is what happened. Generative AI made applying trivially easy. Cover letters in 30 seconds. CVs keyword-tailored on command. Personalised applications at industrial scale. Job seekers used these tools, which is entirely reasonable.

The problem: so did everyone else.

Application volume surged 93% in the past twelve months, per ResumeHog’s May 2026 hiring trends report. Recruiting teams drowned. Companies responded by deploying AI to handle the volume. Today, 87% of companies use AI or automated systems to filter initial applications before a human looks at them.

The conversion rate across all applications: 0.5%.

You are not competing with other candidates when you apply. You are competing with the training data of an algorithm built to replicate past successful hires. If your application looks like the thousands of others it sees daily, it does not pass through. Regardless of your actual ability.

The Three-Signal Strategy

Here is the reframe that changes everything.

In any crowded market, the currency is not volume. It’s specificity. The job market in 2026 is the most crowded in history at the application layer, which means specificity is now the only durable edge.

The candidates getting through are doing three things.

They target at a depth most people skip. Instead of applying to every vaguely relevant role, they identify ten to fifteen companies where their background is not just useful, but specifically rare. There is a difference between “a marketer who has worked in fintech” and “a demand generation lead who built and owned a B2B pipeline from scratch at a VC-backed fintech after Series A.” The second person is not just qualified. They’re specific. Specificity is what passes through AI screens, and what catches a recruiter’s eye in the 25% of applications that actually reach a human.

They deliberately resist AI polish. Between 40% and 80% of applicants now use AI to generate their materials, per research from WasItAIGenerated. The pattern is detectable by human recruiters and increasingly by the AI screeners themselves. Applications that read like a real person wrote them, with genuine specificity, real personality, and the kind of rough edges that signal actual thinking, now stand out against a backdrop of synthetic smoothness.

They demonstrate AI literacy, not just AI usage. Salesforce CEO Marc Benioff said in May 2026 that his company is “keeping engineering lean thanks to AI, but actively hiring in sales.” The pattern across multiple sectors is the same. Companies are not hiring people who use AI. They’re hiring people who understand when and how to deploy it in ways that actually produce output. That distinction is visible in an interview. It’s also visible in a well-written CV.

“There are basically two ways to know you have a future in this market: one, you have vocational skills that AI cannot replicate. Or two, you think in ways that AI tools genuinely cannot emulate.”

Alex Karp, CEO of Palantir

Karp’s framing is sharper than it appears. He’s not saying you need to be a plumber or a prodigy. He’s saying the people who keep thriving bring something an algorithm cannot reproduce: physical craft, creative judgment, or the kind of unusual pattern recognition that makes you genuinely difficult to assess with a keyword filter. That quality does not show up in a synthetic cover letter. It shows up in how you write and how you present yourself when given the chance.

Where you apply matters as much as how you apply

Here is the piece most job hunting advice misses entirely.

If 75% of CVs never reach a human, the highest-return move is not improving your application materials. It’s choosing where you apply.

Mass job boards aggregate volume, which is exactly why they attract algorithmic screeners. The platforms where humans still read applications first tend to be curated, specialised, or invitation-based. The vetting happens at entry, which means if you get access, you’re already past the first filter.

This is why thefirsthire.com is worth paying attention to. Rather than listing every available role, The First Hire connects breakout startups (companies with significant funding, proven founders, and genuine equity upside) directly with candidates in Product, Engineering, Marketing, and Operations. The companies on that platform are not processing 500 applications through an algorithm. They’re looking for specific people. That changes the entire dynamic of the search.

If you have the specificity to target well, you want to be where the vetting has already happened on the company side too.

The real risk in 2026 is passivity

Goldman Sachs estimates AI is displacing 16,000 US jobs monthly in 2026. The World Economic Forum projects 92 million jobs will be displaced before the 170 million new roles materialise. The gap between displacement and creation is where most careers either adapt or stall.

The ones who adapt are not always the ones with the best credentials. They’re the ones who recognised early that the old playbook stopped working, and built a new one.

PwC’s 2026 Global AI Jobs Barometer found that workers in AI-exposed roles who actively developed complementary skills were earning wages 25% higher than those who didn’t. The market is not punishing people for being in AI-adjacent sectors. It’s punishing passivity within them.

The job market in 2026 is not harder for everyone. It’s harder for people who haven’t updated their approach. For those who have, the market is more navigable than the headlines suggest.

What’s the biggest change you’ve made to how you job hunt since AI entered the picture? Hit reply. I’d genuinely like to know.

Until next time, target smart.

- Rayn

P.S. If you want to think through your positioning in this market, whether you’re searching for your next role, trying to attract better candidates, or figuring out which direction to move, I do 1:1 sessions at topmate.io/narayanan.

TLDR

• Overall GTM hiring fell 15% in Q1 2026, but AI-native startups accelerated hiring by nearly 50% year-on-year

• AI companies doubled SDR headcount while selling the very tools that supposedly replace SDRs

• The GTM Engineer role has doubled in a year - 400+ at US digital natives, with total comp reaching $250K+ at Vercel and OpenAI

• 3-in-5 open GTM roles today are AE or Solutions Engineer positions - buyers still want a human for complex deals

• The best startup GTM roles are rarely posted; most are filled through backchannels before the job board listing ever goes live

The market headline is misleading

The number going around: 22,988 GTM job posts in Q1 2026 at US digital native companies, down 15% from Q1 2025. Kyle Poyar published the H1 2026 State of GTM Hiring report through Growth Unhinged last month, built on real-time job data from Sumble, and the top line is not pretty.

SDR roles specifically are down 21% year-on-year.

BUT.

The companies most associated with replacing GTM jobs are also the ones hiring into those same roles most aggressively. Cursor is hiring SDRs. So is Decagon. So is OpenAI. While the broader market pulled back on SDR hiring by 21%, AI-native startups more than doubled their SDR headcount year-on-year, per the same report.

The companies selling the automation apparently are not buying all of it and that’s the fact no one can openly deny.

The Three-Lane Split: where GTM careers actually go from here

The GTM job market has separated into three lanes. If you’re not thinking about which lane fits you, you’re probably applying to the wrong companies.

Lane 1: AI-native SDR

If you’re an SDR, the worst move right now is continuing to target traditional SaaS companies that are cutting headcount. The better move is to go where the hiring is actually accelerating.

Legora, LangChain, Fireworks AI, Decagon, OpenAI. These companies are hiring SDRs at a rate roughly 50% higher than the overall market, according to Poyar’s data. The role itself is evolving too. The SDR job at an AI company in 2026 looks less like high-volume cold calls and more like orchestrating AI agents to handle the data-heavy prospecting while you focus on the judgment calls that require a human.

This is the role sometimes called the “Strat-Agent”: a small team of technical strategists managing AI-driven outbound systems, handling what the agents can’t - political alignment within target accounts, relationship architecture, the final push to close.

Lane 2: GTM Engineer

This is the fastest-growing role in revenue right now. GTM engineering headcount has doubled in a year. There are 400+ GTM engineers at US digital natives, with LinkedIn listing over 3,000 open positions.

Total compensation at leading companies has reached $252K at Vercel, $250K at OpenAI, and $184K at Ramp.

A GTM Engineer builds the systems that make outbound scale. Half commercial thinker, half builder. The skills companies are filtering for: SQL fluency, Python at script level, Clay or n8n at expert level, CRM architecture (HubSpot or Salesforce), and at least one LLM-in-the-loop workflow that shipped and survived three months in production.

You don’t need to be a software engineer. You need to be technical enough to build the machines and commercial enough to know what they should do.

Lane 3: AE and Solutions Engineer

3-in-5 open GTM roles today are AE or Solutions Engineer positions, per the same report. Complex enterprise deals still require a human in the room. This lane has not contracted the way SDR roles overall have.

The shift here is that AI-forward companies now expect their AEs to use AI tooling natively, not treat it as optional extra credit.

The Focus Statement that gets you in before the posting goes live

The best startup GTM roles are rarely posted publicly especially never on LinkedIn Job Posting. By the time a job appears on a board, the best candidates are already in conversation with the founder.

“Founders aren’t looking for generic talent; they are looking for extremely specific experience, and just as importantly, culture fit. Someone who has sold to an exact buyer, operated at this exact stage, and matches their company dynamic.”

- TalentPluto, 2026

That means your positioning matters more than your experience.

Write a Focus Statement: one short paragraph you can send to any warm contact. It should include the role(s) you’re targeting, the stage and GTM motion you’re best suited for (PLG, founder-led, enterprise), and a shortlist of 5 to 10 companies you’re genuinely excited about.

Not “open to opportunities.” An exact brief. Something a founder or a trusted contact can match against a need the moment they hear about it.

The Clay Community Slack has 20,000+ members, many of them deep in GTM engineering and active in the conversation about who’s hiring. Your past managers, alumni networks, and operator communities are others. Get your Focus Statement into those conversations now, before you touch a job board.

I found this job board that’s getting waves in the GTM communities, check it out here,

The First Hire

The number that tells you where to aim

ICONIQ Growth tracked 150+ B2B software GTM leaders and found that AI-forward companies generate roughly $640K net new ARR per GTM FTE, compared to $370K at lower-adoption peers - nearly double.

Read that as a targeting filter. The companies running lean and AI-first are generating more output per person and have the unit economics to keep hiring. The companies still building the way they built in 2022 are the ones pulling back.

Don’t target the GTM job market broadly. Target the companies in Lane 1 and 2 specifically. Write your Focus Statement. Get into the backchannels early.

The roles worth having rarely wait for you to find the listing.

If you’re currently looking for a GTM role at a startup, what’s the actual blocker - knowing which companies to target, or getting in front of the right people? Hit reply and tell me.

Until next time, keep building.

- Rayn

Working on your next move in GTM?

I do 1:1 coaching for operators and founders navigating career transitions, early-stage GTM strategy, and building revenue motion from scratch. If you want to talk through where you are and what the right next step looks like for you specifically, book a session at topmate.io/narayanan.

Here’s a more honest approach: instead of asking “what should I do?”, ask “what should I stop doing?” That one flip changes everything.

The inversion framework and why it works

Charlie Munger, the late Berkshire Hathaway vice chairman, had a principle he borrowed from the mathematician Carl Jacobi: invert, always invert. The idea is simple. When you can’t figure out how to succeed at something, map out all the ways to fail at it instead. Then avoid those things rigorously.

It sounds obvious. It’s surprisingly hard to actually do, because our instinct when we’re lost is to seek more options, not fewer. We want a roadmap. We want someone to tell us the answer. But more options, more information, and more advice often just compound the paralysis.

What actually helps is getting clear on what you’re willing to rule out.

A McKinsey research note on decision quality found that the biggest drag on good decision-making isn’t lack of information. It’s the emotional state people are in when they make decisions. Which brings us to the list.

The behaviours that keep you stuck

When you’re feeling lost, certain habits feel like they’re helping. They’re not. Here are the ones worth watching for:

Making emotional decisions. When you feel lost, there’s a pressure that sets in, the sense that time is running out and you need to act now. That urgency is almost always false. Most decisions have more time than they feel like they do. If you need more time to think, take it.

Blaming external circumstances. The economy, your manager, the industry, your timing. Blame feels useful because it locates the problem somewhere you can’t reach. But if the problem is out there, so is the solution, and that means you’re stuck waiting. The moment you stop blaming is the moment you actually have some leverage.

Numbing the discomfort. Binge watching. Mindless scrolling. Impulse spending. These don’t resolve the feeling of being lost, they just postpone it. And they tend to compound the problem because you’re now a day further along and the discomfort is still there.

Procrastinating on the thing that actually matters. Not chasing what you want is the easiest path available. Millions of people take it. It’s dressed up in very reasonable-sounding language: “I’m not ready yet.” “The timing isn’t right.” “I’ll start next month.” At some point, you have to see it for what it is.

What to do instead

Confront the discomfort rather than running from it. This isn’t motivational advice, it’s practical. The feeling of being lost is carrying information. It’s telling you something is misaligned, something needs attention. If you push it away, you lose access to that signal.

And then, act. Not a big dramatic pivot. Just one small thing today that points in the right direction. Research from Stanford’s Behaviour Design Lab shows that the smallest possible version of an action, taken consistently, builds more momentum than large sporadic efforts. You don’t need a plan. You need a start.

“Life is either a daring adventure or nothing at all.” — Helen Keller

That line is old but it holds. Being lost isn’t a failure state. It’s part of the journey. Every operator, founder, and professional who’s built something meaningful has spent time in this feeling. The ones who came out of it weren’t the ones with the best roadmap. They were the ones who stopped running from the discomfort long enough to figure out what they actually wanted.

🚀 YC Startup Spotlight

Reflect (YC W24)

What it does: A networked note-taking tool that helps you think more clearly by connecting your ideas, decisions, and reading across time. It’s built around how the brain actually works, not how folders work.

Use it when: You’re in a period of career or strategic uncertainty and need a place to think out loud, map your reasoning, and spot patterns in your own thinking across weeks and months.

What’s one decision you’ve been putting off because it felt too big? Hit reply.

Until next time, take care of yourself. Connect with me on LinkedIn

Rayn

The AI SDR market is projected to hit $15 billion by 2030. Hundreds of teams are replacing their sales development reps with autonomous agents. LinkedIn is full of founders announcing they “10x’d their outbound” overnight.

And yet, the number that nobody is talking about: AI SDR tools are churning at 50–70% annually, roughly double the turnover rate of the human reps they were supposed to replace.

Today I want to break down exactly what’s happening, why the companies rushing to automate their outbound are about to get a nasty surprise, and what the 83% of teams actually winning with AI are doing differently.

PRESENTED BY SCRIPTBEE

If your team is still doing outbound the old way: cold lists, generic sequences, and hope, you are competing with companies that have already automated the boring parts and given their reps the signal they actually need.

Scriptbee is the end-to-end B2B revenue platform built for this exact moment. AI co-workers drive high-intent inbound through SEO, AEO, and GEO, so your best prospects are already looking for you. Then Scriptbee de-anonymises your website visitors, enriches decision-maker contacts, and runs automated outbound sequences from first signal to booked call.

No spray and pray. No wasted rep time on cold accounts that were never going to close.

Trusted by 200+ B2B revenue teams globally. If you are serious about outbound in 2026, this is where you start: scriptbee.ai

The number that should worry every revenue leader right now

Let me give you the full picture in three data points.

Eighty-three percent of sales teams using AI saw revenue growth last year, versus 66% of teams without it, according to a 2026 meta-analysis of over 20 studies on AI in B2B sales. AI-powered teams are closing 29% larger deals and running 36% shorter sales cycles. That part you have probably seen in your feed.

Here is the part you have not. According to Autobound’s 2026 State of AI Sales Prospecting report, AI SDR tools are churning at 50–70% annually, roughly double the turnover rate of the human reps they were supposed to replace. And 22% of B2B teams have already fully replaced their SDRs with autonomous agents, meaning a significant chunk of the market is cycling through platforms the way they used to cycle through junior reps who burned out after six months.

Meanwhile, Gartner published a prediction that over 40% of agentic AI projects will be cancelled by end of 2027, not because the technology does not work, but because most teams are implementing it in fundamentally the wrong direction.

So what is actually separating the teams winning from the teams chasing their tails?

The replacement vs multiplier model: why it changes everything

There are two fundamentally different ways to deploy AI in your sales motion. Most teams default to one without realising it.

Replacement Mode means you hand the work to an AI agent and step back. The agent prospects, writes copy, sends sequences, follows up. You fire or stop hiring SDRs. Automation does the job.

It is easy to see why this is tempting. SDRs are expensive, turnover is brutal, and the pitch from AI vendors is compelling: “infinite reps, no salary, no sick days.” When the CFO is asking you to do more with less, Replacement Mode looks like the obvious answer.

Multiplier Mode means your human reps still own the relationship and the judgement. AI handles research, signal monitoring, first-draft personalisation, and prioritisation. The rep approves, refines, and sends. The human is still in the loop, operating at 5–6x the output they could achieve alone.

The data is unambiguous about which works. The highest-performing approach in every meta-analysis of AI in B2B sales is human-in-the-loop. The teams replacing humans wholesale are the ones with 70% tool churn and abandoned projects. The teams treating AI as a force multiplier are the ones with 36% shorter cycles and 29% larger deals.

This is not a capabilities problem. AI can write emails. AI can scrape data. AI can send sequences at 3am.

The problem is that good outbound is not a volume problem. It is a precision problem. And precision requires judgement.

Why automation at scale amplifies the wrong thing

When you automate mediocre outbound, you do not get better outbound. You get mediocre outbound at 10,000 messages a day.

I have seen this play out at Scriptbee. When teams come to us having already burned through one or two AI SDR tools, the pattern is almost always the same: they automated their sequences before fixing their targeting. They had the wrong ICP, the wrong messaging, or both, and the AI made sure that problem was felt by every prospect on their list at once.

“Most agentic AI projects right now are early stage experiments or proof of concepts that are mostly driven by hype and are often misapplied.”

Anushree Verma, Senior Director Analyst, Gartner

The Gartner framing is right, but I would add one thing: the misapplication is not random. It is structural. Companies under pressure to show outbound results reach for the fastest-looking lever (autonomous AI) rather than doing the slower work of getting their signal stack right first.

What the 83% actually did differently

The teams seeing real results from AI in sales followed a simple sequence, even if they did not name it explicitly.

They fixed the inputs before they automated the outputs.

That means: clean ICP, validated messaging, real intent signals (who is visiting your site, what they looked at, how long they stayed), and enriched decision-maker data, all before any sequence gets deployed. The AI then operates on high-quality signal rather than a cold list. The difference between those two scenarios is not the AI. It is what the AI is working with.

From there, the human rep’s role shifts entirely. They stop spending 70% of their day on research and list-building. They start spending 70% of their day on the calls, the demos, and the relationships that only a human can run well. The AI earns the meeting. The rep closes it.

This is not a hypothetical. At Scriptbee, the teams getting the best results are the ones who came to us after burning through autonomous SDR tools. Not because they needed more automation, but because they finally understood what they were actually trying to automate.

That is the Replacement–Multiplier distinction in practice. The question is not “AI or human.” It is: where does human judgement sit in the workflow, and what exactly are we using AI to replace versus amplify?

A lot of teams never ask that second question. They skip straight to the tool.

How is your team currently using AI in your sales motion: as a replacement or as a multiplier? Hit reply. I genuinely want to know where you are at.

Working with me 1:1

If your outbound is producing noise but not pipeline, we should talk.

The frameworks in this issue are exactly what I cover in my 1:1 coaching sessions, working through your ICP, your signal stack, and your AI implementation so the work compounds rather than churns. I have worked through this with early-stage founders and heads of revenue at $10M+ ARR companies. If that sounds like your situation, book a session: topmate.io/narayanan

Until next time. Keep your humans where they matter most.

Rayn

Connect with me on LinkedIn → linkedin.com/in/chalkmeout

Here’s the thing: your mood isn’t something that just happens to you. It’s something you can influence, if you’re willing to pay attention.

The first step is awareness, not control

Most people assume bad moods are caused by outside events. The difficult client, the failed campaign, the deal that fell through. And sure, those things trigger it. But staying in a bad mood? That’s on you.

The Stoics had a version of this. You can’t control what happens. You can only control your response to it. That’s easy to say and genuinely hard to do, especially when you’re running a business and the stakes feel real every single day.

But here’s a practical place to start: for the next two weeks, log your mood like you’d log your energy levels or your workouts. Not to judge yourself, just to notice. What triggers it? Time of day? Certain people? Certain types of tasks? You might be surprised how predictable your bad moods actually are once you start tracking them.

Research from the American Psychological Association consistently shows that emotional awareness, simply naming and recognising an emotional state, is one of the most effective ways to reduce its intensity. You don’t need to fix the mood immediately. You just need to see it clearly.

The “do nothing” move that actually works

When something happens that puts you in a bad mood, your first instinct is almost always the wrong one.

Someone sends you a passive-aggressive message. A team member drops the ball publicly. A competitor takes a dig at you online. Your gut says: respond now.

Don’t.

Take ten seconds. Look at what’s actually at stake. If your integrity or an important relationship is on the line, yes, act. But if it’s noise dressed up as urgency? Let it pass. The response you write in a bad mood is almost never the one you’d be proud of an hour later.

This isn’t about being passive. It’s about being deliberate. The operators who manage their mood well aren’t the ones who feel less. They’re the ones who pause long enough to choose their response.

You can generate a good mood, not just wait for one

This is the part most people miss. Managing your mood isn’t only about avoiding the bad stuff. It’s about actively creating the conditions for a better state.

A few things that actually work: physical movement (even a 10-minute walk shifts things meaningfully), deliberate context switching, and what some psychologists call “broaden-and-build,” stepping back to see the bigger picture when you’re locked in a spiral.

The goal isn’t to never feel bad. Bad moods carry real information. They tell you something is off, something needs attention, a boundary has been crossed. The goal is to not stay there longer than necessary.

Set a rough internal rule for yourself: 10 minutes of sitting with it is fair. After that, move. Not by suppressing it, but by redirecting your attention to something that matters.

A study published in Cognition and Emotion found that people who use reappraisal strategies, actively reframing a situation rather than just venting or suppressing, report higher wellbeing and better decision-making over time. As an operator, your decision-making is your product. Your mood is directly upstream of it.

🚀 YC Startup Spotlight

Focus Buddy (YC W25)

What it does: An AI productivity coach that stays on a call with you while you work, helping you stay focused, recover from distractions, and avoid burnout. It detects when you’re drifting and intervenes in real time through voice.

Use it when: You’re working through a high-stakes focused block and need something that holds you accountable without the friction of another app to manage.

What triggers your bad mood most often at work? Hit reply, I’m genuinely curious.

Until next time, take care of yourself.

Connect with me on LinkedIn

Rayn

I know it sounds almost too simple. But spend time with an 80-year-old who’s still building things, still learning, still picking up the phone, and you’ll feel the difference immediately. There’s an aliveness to them that has nothing to do with their energy levels and everything to do with their orientation toward the world.

Curiosity isn’t soft. It’s a strategy.

There’s a tendency to treat curiosity as a personality trait, something you either have or you don’t. But that’s not how it works.

Curiosity is a practice, a daily habit. And like any practice, it atrophies when you don’t use it.

This is what tends to kill curiosity: busyness. When you’re fully maxed out (back-to-back meetings, a full inbox, a to-do list that breeds overnight) there’s no space left to wonder about anything. You’re in execution mode constantly. And execution mode, by design, only looks at what’s directly ahead.

The problem isn’t that busy people are incurious. The problem is that busyness creates an illusion of forward motion while actually shrinking your world. You’re moving fast inside a smaller and smaller circle.

The operators and entrepreneurs who tend to thrive over decades, not just for a year or a funding cycle, are almost always the ones who protect some space for curiosity. They read things outside their category. They ask questions they don’t need the answers to. They follow threads that go nowhere obvious.

What “staying curious” actually looks like in practice

It doesn’t mean reading 50 books a year or signing up for online courses you never finish.

It looks more like this: the 82-year-old entrepreneur who flew to Amsterdam by himself, still working on new tech, still genuinely excited about what’s next. The investor in his seventies who has never wanted to retire, not because he needs the money, but because stepping back feels like giving up the thing he loves most.

What these people share isn’t a schedule hack. It’s a refusal to close off. They treat every new conversation, every new problem, every new industry development as something worth understanding, not just something to react to.

For operators, this matters more than it might seem. The market shifts. Tools change. What worked last year stops working. The ones who adapt fastest aren’t always the most experienced. They’re often the most curious. They didn’t stop asking “why does this work?” when they found something that worked.

A World Economic Forum report on future skills consistently puts curiosity and continuous learning near the top of the traits employers and leaders will need through 2030 and beyond. This isn’t a soft skill anymore. It’s becoming a core competitive advantage.

The relationship between curiosity and longevity

There’s actual science behind this, not just anecdote.

Research on cognitive health suggests that people who remain mentally engaged, pursuing new knowledge, new skills, new social connections, show slower cognitive decline as they age. The brain, much like a muscle, responds to use. Curiosity, in this sense, is a form of maintenance.

But more immediately practical for founders and operators: curiosity is what keeps you from becoming the person who’s always fighting yesterday’s battle. Every industry eventually produces people who “know how things work” and then, at some point, can’t figure out why things stopped working. That’s usually not a market problem. It’s a curiosity problem.

“Everything is noise except for doing the things that you’re curious about.”

That line stuck with me. Because it reframes curiosity not as a nice-to-have, but as a filter. Not everything deserves your attention. But the things you’re genuinely curious about? Those deserve it completely.

🚀 YC Startup Spotlight

Read Bean (YC W24)

What it does: Helps intermediate and advanced language learners improve reading skills using real-world content (articles, news, native texts) adapted to your level. It measures where you are, transforms content into bite-sized lessons, and reinforces vocabulary over time.

Use it when: You’re trying to learn a language as an adult and the usual apps feel too juvenile. Real content, adapted to you. That’s the curiosity-friendly way to learn.

What’s one thing you’ve been genuinely curious about lately that has nothing to do with your immediate work? I’d love to hear it. Hit reply.

Until next time, take care of yourself.

Connect with me on LinkedIn

Rayn

That’s not me being optimistic, I run a software company too. That’s Mike Cannon-Brooks — co-founder of Atlassian, a company doing north of $5 billion in cloud revenue and accelerating responding to the loudest narrative in tech right now. And honestly, he’s right. Software isn’t going anywhere, well I hope so. But a lot of software companies are about to have a very rough few years.

This is what is actually happening, and what operators need to understand about which side of the line they’re on.

The “Death of SaaS” panic is a Category Error

Every few weeks, someone on X declares that SaaS is dead, AI ate it, and we should all pivot to agent wrappers. Then the same week, a company like Harvey raises $200 million at an $11 billion valuation. Harvey is… software. Growing 300% year-over-year.

So which is it?

On a recent episode of 20VC, Cannon-Brooks broke it down simply: will every SaaS company survive the next five to ten years? Absolutely not. Will a lot of them continue to grow and prosper? Absolutely. Is that any different from the last decade? No.

He pulled up Atlassian’s old competitor lists from 2005, 2010, 2015. Most of those companies don’t exist anymore — merged, acquired, or gone. And Atlassian has a completely new set of competitors today. That’s just how the technology industry works.

The difference now is the rate of change. In a normal cycle, maybe 5-10% of companies fail each year. During an architectural shift — cloud, mobile, and now AI — that death rate compresses into a two-year window where it could hit 50%. Not every company dies, but the ones that aren’t rebuilding fast enough absolutely will.

Software customers are going to buy more in the future. The question is whether they’ll buy it from you.

The input-output framework that changes everything

Here’s the most useful mental model from the conversation, and it comes directly from how Cannon-Brooks thinks about AI’s impact across different business functions.

Some domains are input-constrained. Legal, customer support, HR — there’s a fixed number of problems coming in. Your lawyers can’t create more legal problems just because they got faster. Your customers ask 100 questions a day, and if you have twice as many customers, that becomes 200. The input is based on some external ratio.

Other domains are output-constrained. Engineering, product development, creative work — the roadmap is never finished. You can always build more. The constraint is how much you can produce, not how much demand exists.

Here’s why this matters for operators:

In input-constrained domains, AI mostly cuts costs. You answer the same questions with fewer people. That’s a real, measurable ROI — easy to sell, easy to prove. But in year two and three, those savings are baked in. You need to keep delivering new value or you’re just a line item that shrinks.

In output-constrained domains, AI expands what’s possible. Engineering teams aren’t getting smaller — they’re shipping more. According to Cannon-Brooks, Atlassian’s 10,000 R&D people are building things faster and better than ever. Some of their AI features now run at a thousandth of the cost compared to when they launched. The features still work. The margin just went up.

If you’re building a company, ask yourself: am I selling into an input-constrained or output-constrained domain? The answer changes your entire strategy.

The Danger: Being third in a subsegment

The customer support space tells this story perfectly. Sierra just crossed $150 million in ARR. Fourteen new companies in the space have raised over $100 million in the last two years. And they’re competing against ServiceNow, Salesforce, Atlassian, Zendesk, and Intercom.

The market is real. The money is flowing. But as the 20VC crew pointed out, the worst position to be in is third place in a subsegment. You’re too small to dominate, too big to pivot, and the market leader can decide to annex your space anytime they want.

This is happening everywhere. VCs are piling into consensus winners because it’s the venture capital equivalent of “you can’t get fired for buying IBM.” Of newly minted unicorns in Q1 of last year, 40% already had one or more up rounds by Q4. The money says: once you’re winning, we’ll double and triple down.

For operators, the lesson is uncomfortable but clear. If you’re not the obvious leader in your space, you need a different playbook. Adjacent territory. A wedge nobody else owns. Or the honest conversation about whether you’re building something defensible at all.

🚀 Tools of the week

• 🚀 Cloudera: Your weekly playbook for navigating agent chaos and turning AI disruption into a scalable advantage. Listen to The AI Forecast podcast.

• ✂️ EZTrimmer: Trim, split, and export videos in seconds with AI.

• ✅ Notion AI: Use AI directly inside Notion to boost your productivity.

• 📽️ Kubrix: Create cinema-quality stunning AI-generated videos in seconds.

• 🤖 Lorka: Chat with the top AI models in one place.

🧪 YC Startup Spotlight

Abundant (YC W25)
• Provides on-demand human operators who step in when AI agents fail in production, ensuring 100% reliability while generating training data for improvement.

• Use when: You’re deploying AI agents in production and need a safety net for edge cases, especially in regulated industries like healthcare, legal, or finance. abundant.ai

Until next time, keep building, keep questioning, and don’t let the noise drown out what actually matters.

- Rayn

Bill Gross, the founder of Idealab, studied hundreds of startups and found that timing accounted for 42% of the difference between success and failure. The idea — the thing most founders obsess over — came in third.

Sam Altman puts it differently: wait to have a good idea before you start a startup. If you start without one and just cast around, you’ll be under pressure to make something up. And made-up ideas don’t survive contact with reality.

So how do you find the right idea at the right time? Here’s what the data and the best founders actually tell us.

Presented by Scriptbee

If you’re building a startup in 2026, here’s a question worth asking: when someone asks ChatGPT or Perplexity for a solution in your category, does your brand show up?

Most founders have no idea. Scriptbee gives you the answer. Track your brand’s visibility across AI search engines, see which queries trigger mentions of your product, and optimise your content to win more AI-generated recommendations. One B2B SaaS company grew their AI visibility by 40% in 60 days.

WHAT MATTERS TODAY

1. The clone trap kills more startups than bad execution

Here’s the pattern. Something works — a new app, a business model, a category — and thousands of people rush to copy it. It happened with social networks after Facebook. It happened with photo-sharing apps after Instagram. And it’s happening right now with AI wrappers.

Altman shares a revealing stat from when he joined the board of Helion, a nuclear fusion company: that same year, roughly 10,000 photo-related startups launched alongside it. One nuclear fusion company. Ten thousand photo apps.

The odds were obvious, but nobody was paying attention. Meanwhile, 42% of startups fail because they build products nobody actually wants — the single biggest killer. Not bad code. Not weak marketing. Building something the market never asked for.

The takeaway for operators: if you can name ten companies doing the same thing you’re doing, that’s not validation — that’s a warning sign.

2. The best ideas come from personal pain, not market research decks

Strip away the mythology and most transformational companies started with someone solving their own problem. Altman is emphatic about this — a very high percentage of the best startups trace back to founders scratching their own itch.

Airbnb’s founders needed to pay rent. They noticed hotels in San Francisco were sold out during a design conference, so they put air mattresses in their living room. That wasn’t a market analysis. That was survival. Stripe’s founders were building websites and kept running into the same nightmare: payments were absurdly hard to integrate. They didn’t conduct a survey. They felt the pain every single day.

This matters because personal experience gives you an unfair advantage. You understand the nuance. You know what “good enough” actually looks like to the end user. You don’t need a focus group to tell you the problem is real — you’ve lived it.

The data backs this up. First-time founders have an 18% success rate. Those who’ve previously failed do slightly better at 20%. But founders with a successful exit? They hit 30%. Why? Because they’ve developed the pattern recognition to spot real problems versus imagined ones.

3. Timing is the invisible advantage nobody talks about enough

Here’s where Altman’s thinking and Gross’s data converge beautifully. Both argue that great companies cluster around major technological waves — not because founders coordinate, but because the wave creates new problems that need solving.

The internet wave in the mid-to-late 90s produced Amazon, Google, and Yahoo. The mobile wave after the iPhone in 2007-2008 produced Uber, Airbnb, Instagram, and WhatsApp. Each cluster appeared in a tight window, right after a major platform shift.

Altman’s insight is that you don’t need to create the wave — that takes too many resources for a startup. What you can do is notice it early, while others still think it’s a toy, and build on top of it.

The current wave? AI. But here’s the nuance that separates winners from the wreckage: almost $200 billion was poured into AI startups in 2025 alone, yet 95% of generative AI pilot projects in enterprises failed to deliver measurable ROI. The money is flowing, but the real problems — the ones where AI genuinely changes the outcome — are still being discovered.

The areas still underbuilt right now? Product discovery (figuring out what to build, not just coding it). AI agent reliability in production environments. AI-native companies that don’t just sell tools to incumbents, but replace the incumbents entirely — what YC calls “full-stack AI.”

The best founders don’t create waves. They notice them while everyone else still thinks they’re a toy.

QUICK WINS

🎯 Test your idea with one question: “Is this my problem?” If you haven’t personally felt the pain, you’re guessing. Guessing gets expensive fast.

🔧 Use the “10,000 photo apps” filter: Before building, search Product Hunt, YC’s directory, and Crunchbase for competitors. If you find dozens doing the same thing, either find a genuinely different angle or move on.

💡 Time your wave, not your launch date: 10% of startups fail specifically because of mistiming — launching too early or too late. The question isn’t “is this a good idea?” It’s “is the market ready for this right now?”

🏄 Look where the money’s going, then go adjacent: The crowded plays in AI are chatbots and copilots. The underbuilt opportunities are in infrastructure, reliability, and vertical AI for industries that haven’t been touched yet — construction, logistics, legal, agriculture.

📊 Talk to 20 potential users before writing a single line of code: Startups with mentors are 3x more likely to succeed. But even before mentors, the cheapest insurance is simply asking people whether they’d pay for what you’re building.

Don’t start a startup without a good idea. Start with the problem — the idea will follow.

Looking to make sharper decisions faster?

If today’s piece hit a nerve; if you’re wrestling with which idea to pursue, or whether the timing is right. I work 1:1 with founders and operators to stress-test ideas and build systems that stick. Book a session: https://topmate.io/chalkmeout

🚀 YC Startup Spotlight

Conveo (YC S24)

An AI-powered qualitative research platform that conducts and analyses voice and video interviews at scale — delivering customer insights in hours, not weeks. Trusted by Unilever, Google, and Sanofi, Conveo raised $5.3M in seed funding to help teams understand what customers actually want before building.

Use when: You’re validating a product idea or testing positioning and need deep customer insights without the 6-week agency timeline. conveo.ai

That’s all for today. If something here resonated, forward this to a founder friend who’s still searching for “the idea.” Sometimes the nudge matters more than the advice.

Connect with me on LinkedIn

Until next time,
Rayn

OpenAI started putting ads inside ChatGPT. A Harvard study confirmed that AI is making us work more, not less. And Amazon quietly started building a marketplace where publishers can sell their content directly to AI companies.

If you zoom out, the pattern is obvious: AI is no longer a tool you choose to use. It’s becoming the environment you operate in. The ads you see. The pace you work at. The way your content gets monetised. All of it is shifting — fast.

Here’s what actually matters for operators this week.

PRESENTED BY Scriptbee

If ChatGPT is now running ads inside conversations, the question for your brand isn’t just “are we advertising there?” — it’s “are we even showing up when someone asks AI about our category?”

Scriptbee tracks your brand’s visibility across ChatGPT, Perplexity, Claude, and Gemini. See which queries trigger your citations, where competitors are outranking you, and what to fix. One B2B SaaS company grew AI visibility by 40% in 60 days.

→ See where your brand stands in AI search

WHAT MATTERS TODAY

1. ChatGPT now has ads — and it changes the game for AI search

OpenAI officially flipped the switch on Monday. Ads are now live for Free and Go tier users in the US — appearing at the bottom of ChatGPT responses, clearly labelled and visually separated from the AI’s answers.

The timing is worth noting. This launched literally one day after Anthropic ran Super Bowl ads mocking the entire concept of AI chatbot advertising. OpenAI’s response? Press ahead anyway.

Here’s what operators should pay attention to: ads are matched based on your conversation topic, past chats, and previous ad interactions. If someone asks ChatGPT for project management tools and your competitor has bought that placement — they’re showing up, not you.

The operator takeaway: ChatGPT isn’t just a productivity tool anymore. It’s becoming an advertising platform. And if your brand isn’t visible in AI conversations organically or through paid placement, you’re ceding territory to competitors who are.

2. AI doesn’t save you time — it makes you do more

Researchers from UC Berkeley spent eight months embedded inside a 200-person tech company, studying how AI tools changed daily work. Their findings, published in Harvard Business Review, are a reality check for every founder selling “10x productivity” narratives.

What they found: employees didn’t use AI to finish early. They used it to take on more. Product managers started coding. Researchers took on engineering tasks. People fed work to AI during meetings, breaks, and lunch. Nobody asked them to — they just did.

The researchers identified a dangerous pattern they call “workload creep.” Burnout was reported by 62% of associates and entry-level workers, compared to just 38% of C-suite leaders. As one engineer put it: “You don’t work less. You just work the same amount or even more.”

The operator takeaway: if you’re rolling out AI tools across your team, don’t just measure output. Measure intensity. Build in what the researchers call “intentional pauses” — structured breaks, sequenced workflows, and protected time for human collaboration. The companies that win with AI won’t be the ones that adopt fastest. They’ll be the ones that adopt most thoughtfully.

3. Amazon and Microsoft are building AI content marketplaces — and it matters for every creator

According to The Information and TechCrunch, Amazon is in talks with publishers about launching a marketplace where content owners can license their material directly to AI companies. AWS has already circulated slides grouping the marketplace alongside its core AI tools like Bedrock.

This follows Microsoft’s Publisher Content Marketplace launch last week — a licensing hub that lets publishers set usage terms for AI companies building with their content.

Why this matters for operators: the AI industry has spent the last two years scraping content and negotiating one-off deals with major publishers. Now, we’re moving toward structured marketplaces with usage-based pricing. If you’re creating content — whether that’s a newsletter, a blog, or a knowledge base — your material is becoming a tradeable asset in the AI supply chain.

The operator takeaway: start thinking about your content as a product, not just a marketing channel. The companies and creators who have well-structured, high-quality content libraries will have leverage as these marketplaces scale. This is the beginning of a content licensing economy that could reshape how publishers monetise in the AI era.

QUICK WINS

🎯 ChatGPT ad controls exist: If you’re a Free or Go user, you can dismiss ads, turn off personalisation, and delete your ad data with one tap in Settings. Worth doing now before the defaults settle in.

🔧 NotebookLM now connects to Gemini: You can link your NotebookLM notebooks directly to Gemini’s chat box. Click the (+) icon, select your notebook, and get contextualised answers from your own curated sources. Handy for operators sitting on research libraries.

💡 The “AI practice” framework: The Harvard researchers recommend three things for teams using AI — intentional pauses (scheduled breaks), sequencing (coordinated workflow phases), and human grounding (protected time for dialogue). Simple, but most companies aren’t doing any of them.

🚀 Content licensing is the next SEO: Just as brands once optimised for Google, smart operators will soon optimise for AI training marketplaces. If you’re producing original content at scale, keep your archives clean, well-structured, and rights-clear.

📊 The burnout gap is real: 62% of entry-level workers reported AI-related burnout vs. 38% of executives. If you’re a founder or manager, this gap should worry you — the people doing the heaviest AI-augmented work are the most at risk.

FAQ

Will ChatGPT ads affect the quality of responses? OpenAI says no — ads are displayed separately at the bottom of responses and don’t influence the AI’s answers. Paid tiers (Plus, Pro, Business, Enterprise) won’t see ads at all. But the precedent is set, and sceptics are watching closely.

Does the Harvard study mean AI tools aren’t worth using? Not at all. The research shows AI does increase output. The problem is that without intentional boundaries, workers absorb the gains as extra workload rather than reclaiming time. The fix isn’t less AI — it’s better AI adoption practices.

Should I care about AI content marketplaces if I’m not a publisher? Yes. If you create any content — blog posts, newsletters, documentation, courses — you’re a publisher. These marketplaces could eventually let you license your content to AI companies. Getting your content house in order now is a smart long-term move.

“You don’t work less. You just work the same amount or even more.” — Engineer quoted in the Harvard Business Review study on AI and work intensification

The companies that win with AI won’t be the ones that adopt fastest. They’ll be the ones that adopt most thoughtfully.

Feeling the AI intensity yourself?

You’re not imagining it. The pace is real, and most operators are figuring this out alone. I work 1:1 with founders and operators to build sustainable systems — for AI adoption, for team management, for cutting through the noise.

Book a session: https://topmate.io/chalkmeout

🚀 YC Startup Spotlight

Basepilot (YC W24) • Builds AI coworkers that automate repetitive browser-based tasks — data entry, form filling, lead sourcing — by learning from simple demonstrations. No code required, and teams report saving around 30% of their time weekly on mundane back-office work.

• Use when: Your team is drowning in repetitive workflows across web apps and you want to free up capacity for higher-value work — without hiring or building custom automation.

basepilot.com

What’s one thing you’ve changed about how your team uses AI after noticing the workload creep? Hit reply — I’d genuinely love to hear what’s working (and what’s not).

That’s it for this week’s deep dive.

The AI landscape is moving fast, but that doesn’t mean you have to sprint every second. Sometimes the smartest operator move is to slow down just enough to see where the ground is shifting.

If something here landed for you, forward this to one person who needs to read it. And if you haven’t already, connect with me on LinkedIn — that’s where I share the stuff that doesn’t always make it into the newsletter.

Until next time,

Rayn

They’re not bad at building them. They’re just... slow.

You spend three hours updating your runway forecast. Another two debugging a circular reference. Half a day rebuilding last quarter’s actuals because you changed how you track CAC.

And by the time you’re done, the numbers you wanted to test are already outdated.

I just tried Claude in Excel (Anthropic’s new add-in), and honestly? It solves this exact problem. You chat with it like you’d chat with a CFO who actually gets spreadsheets.

Ask it to explain a formula. Update assumptions without breaking dependencies. Build a three-statement model from scratch. All without leaving Excel.

This isn’t about replacing your finance knowledge. It’s about getting back the 10 hours a week you spend wrestling with formulas instead of analysing scenarios.

This issue is brought to you by Scroll

Your team wastes hours searching for information across Slack, Notion, Google Drive, and other tools.

Scroll.ai connects all your tools and lets your team ask questions in natural language to find what they need instantly. Instead of searching five places, ask Scroll: “What’s our Q4 hiring plan?” and get the answer immediately.

Thousands of companies rely on Scroll to automate knowledge workflows across documentation, RFPs, and agency work. One founder cut their team’s search time by 60%. Check out Scroll.ai.

What is Claude in Excel?

It’s an add-in that puts Anthropic’s Claude Opus 4.5 model directly in your Excel sidebar. You chat with it like you’d chat with a CFO who actually understands spreadsheets.

Ask questions about specific cells. Update assumptions while preserving formulas. Debug errors. Build new models from templates. All without leaving Excel.

According to recent analysis, automation and AI-powered templates have reduced the average time to construct comprehensive financial models by 40% over the last five years. Claude in Excel accelerates this even further by turning natural language into executable spreadsheet actions.

Why this matters for startup founders

Most financial modeling tools fall into two camps: either they’re too basic (glorified calculators) or they require an accounting degree to operate.

Claude in Excel sits in the middle. You can ask “What assumptions drive the revenue forecast in Q3?” and get an answer with citations to the exact cells. Or say “Increase growth rate by 2% and show the impact on terminal value” and watch it update everything correctly.

Research shows that 61% of venture capital firms now explicitly require startups to include financial models within their pitch materials. But here’s the catch: investors want dynamic models with scenario toggles and KPI dashboards, not static spreadsheets.

Claude in Excel lets you build these faster. Much faster.

What it actually does well

Reads and explains complex models: Ask about specific formulas or entire sections. Claude navigates multiple tabs and provides answers with direct citations to cells.

Updates assumptions safely: Change values while maintaining all formula dependencies. Every change gets highlighted with explanations.

Builds and fills templates: Create three-statement models from scratch or populate existing templates with data from uploaded documents (like 10-Ks).

Debugs errors: Identifies sources of #REF!, #VALUE!, or circular references and suggests fixes that maintain spreadsheet integrity.

The catch? It’s currently in beta and only available to Pro, Max, Team, and Enterprise users. Chat history doesn’t save between sessions. And it doesn’t handle advanced features like macros, VBA, or conditional formatting.

What founders are actually doing with it

Based on the documentation and early use cases, founders are using Claude in Excel for three main workflows:

Scenario analysis: “Add a downside case assuming revenue drops 15%” or “Create base, bull, and bear scenarios with different growth assumptions.”

Investor prep: “Build a 3-statement financial model for a SaaS company” or “Fill this DCF template with data from the uploaded 10-K.”

Error hunting: “Find all circular references in this workbook” or “Why is this NPV calculation returning #VALUE?”

One operator shared that they used Claude to build a full financial model for a Series A pitch in under four hours – something that would normally take two days with a CFO consultant.

The bigger shift happening here

This isn’t just about Excel getting smarter. It’s about the entire financial modeling workflow changing.

Startups invest in finance early to make better strategic decisions as they scale and to raise capital on stronger terms. But most founders waste hours each week just maintaining their models instead of using them to make decisions.

CB Insights recently reported that fintech startups are increasingly deploying AI agents to handle complex, repetitive financial tasks. Claude in Excel is part of this shift: moving from spreadsheets as databases to spreadsheets as interfaces.

You don’t need to know every formula. You need to know what questions to ask.

Does Claude understand financial modeling conventions?
Yes, Claude is trained to recognise common financial modeling patterns and industry-standard calculations. However, always verify outputs match your specific methodologies.

Can I use this with sensitive data?
Claude works within your existing security framework. For highly sensitive or regulated data, follow your organisation’s data handling policies.

What if Claude makes a mistake?
Claude highlights all changes to your workbook. Review carefully before saving. You can always undo using Excel’s standard undo function.

Will chat history be saved?
Not currently. Each session starts fresh. However, Anthropic is working to support this in future versions.

“A financial model is the numerical expression of your startup’s goals. The best models aren’t ‘right’ – but the differences between projections and actual results drive insight into your company’s potential.”

Building financial models is one thing. Using them to make better decisions is another.

If you’re struggling with investor-ready projections, scenario planning, or simply making sense of your unit economics, let’s talk. I work with founders on everything from first-time fundraising decks to Series A financial models that actually close rounds.

Book a session at topmate.io/chalkmeout and we’ll build a financial model that tells your story.

YC Startup Spotlight

Crunched (YC F25) – The Excel AI analyst for power users in consulting, investment banking, and private equity. It error-checks workbooks end-to-end, builds models from scratch, and handles grunt work like extracting and linking data from information memorandums to template models. Use it when you need to scale financial modeling operations without hiring junior analysts.

What’s the most time-consuming part of your financial modeling workflow right now?

I’m fascinated by how AI is changing the operator toolkit. Not in a “this will replace everyone” way, but in a “this frees you up to think instead of execute” way.

Claude in Excel is exactly that. It’s not going to build your strategy. But it might save you 10 hours a week so you can actually think about the strategy.

And honestly? That’s the unlock.

Let me know if you try it. I’m curious what workflows you build.

Until next,
Rayn

Connect with me on LinkedIn if you want to chat about financial modeling, fundraising, or just startup operations in general.

They set goals. They feel inspired. And by February, it’s over.

Not because they’re lazy. Not because they lack discipline. But because they’re building on sand instead of structure.

Research from the European Journal of Personality confirms what you’ve probably felt: setting attainable goals is more closely linked to well-being than simply aiming for lofty or vague aspirations BetterUp. The problem isn’t your ambition, it’s your architecture.

This week, I’m sharing 20 evidence-backed techniques to make 2026 different. Not through willpower. Through structure.

The Four Acts of Designing Your Best Year

Most people treat goal-setting like writing a shopping list. They scribble down wishes, tape them to the fridge, and hope motivation shows up.

But motivation is unreliable. Structure is what lasts.

Here’s how to build a year that actually works.

Act One: Get Clear on What Actually Matters

Before you set a single goal, run a regret review. Not to punish yourself but to learn.

Pick your biggest regret from 2025. Just one. The thing that still bugs you when you’re lying awake at 2am. Maybe you didn’t protect your creative time. Maybe you drifted from people who mattered. Maybe you spent 300 hours scrolling instead of building.

Write it down. Then write the lesson and the plan to avoid repeating it in January.

Research on regret shows the worst response is ignoring it. The second worst is wallowing. The best? Using it as instruction.

Next, jump forward to December 31st, 2026. Close your eyes. The thing you cared about most didn’t happen. What went wrong?

This is called a pre-mortem, imagining failure before it happens so you can prevent it. If future-you failed because you never scheduled family time, block 10 sacred family days right now. If you failed because nobody held you accountable, find that person this week.

Finally, choose a theme word for 2026. Not a sentence. One word that captures who you’re becoming.

Simplify. Connect. Build. Ship.

Psychologists call this a self-cue; a simple trigger that snaps your attention back to what matters when you drift.

Act Two: Build Structure That Sets You Free

Studies show that writing down specific goals increases the likelihood of achieving them by 42% compared to simply thinking about them Bristolcreativeindustries. But writing goals isn’t enough. You need structure.

Start by protecting your first hour. Research from the University of London found that multitasking early in the day drops your cognitive performance by the equivalent of 10 IQ points. That’s the difference between great work and forgettable work.

Use that first hour for deep work. Writing. Exercise. Reading something that expands your thinking before the world shrinks it. No email. No phone. No reacting to other people’s priorities.

Then adopt the 2-minute rule: if a task takes 2 minutes or less, do it immediately. Small tasks create invisible cognitive clutter. The 2-minute rule is a fog remover.

Every Friday, run a shutdown ritual. Spend 5 minutes reviewing your week, setting your top three priorities for Monday, and blocking 60-90 minutes for something that matters. Then say out loud: “Shutdown complete.”

Sounds corny. Works brilliantly. Your brain gets cognitive closure, which means it can actually rest.

Finally, every Sunday, run a 15-minute reset. Look at what’s coming. What’s essential? What can wait? What needs breathing room?

People who wrote goals clearly and tracked progress were 33% more likely to achieve substantial outcomes Bristolcreativeindustries. The review is as important as the writing.

Act Three: Build Motivation That Lasts

Here’s where most people stumble. They rely on inspiration. But inspiration is fleeting.

Start with the 85% rule. Research from UC San Diego found that systems learn best when they’re right about 85% of the time. Not 100%, not 50% but 85%. That’s the sweet spot where you’re stretched but not snapped.

Pick one goal. Deliberately dial the difficulty so you’re succeeding 8 or 9 times out of 10. If it’s too easy, increase the challenge. If you’re failing constantly, dial back.

That uncomfortable feeling? That’s not failure. That’s the emotional signature of learning.

Next, track small wins daily. Harvard research shows people feel and perform best on days when they experience progress and even tiny progress. At the end of each day, write down three ways you moved forward. Sent that email. Read three pages. Chose the walk over the scroll.

The act itself matters more than the list. Once you see daily wins, momentum becomes hard to stop.

Act Four: Connect and Renew

Nobody sustains excellence alone.

Build a challenge network with a small group who care enough to tell you uncomfortable truths. Don’t just ask “what do you think?” Ask “what’s one thing I can do better?”

Then curate your circle. You need three people: a challenger who won’t let you coast, a cheerleader who believes in you on bad days, and a coach who’s already played the game you’re trying to play.

Sociologist Nicholas Christakis found that emotions and behaviors spread through social networks up to three degrees. Happiness spreads. So does discipline. And their opposites.

Choose carefully.

Finally, create a to-don’t list. Every quarter, ask: what’s not worth my time? A meeting that adds nothing. A committee you’ve outgrown. A project you’re keeping alive out of habit.

Subtraction frees up time and attention in ways addition never can.

“A year is a long time. Treat it as four shorter chapters each with its own focus. Four 90-day pushes will beat one vague 12-month intention every single time.”

The One Practice That Changes Everything

Here’s the final move: send 26 thank-you notes in 2026. One every two weeks.

Research on gratitude shows that written letters produce a lasting boost in happiness, reduce stress, and deepen connection. In one study, gratitude writers reported 25% more life satisfaction and even exercised an extra 90 minutes per week.

Handwritten is best. Short is fine. Simple, clear, sincere.

You’ll brighten someone’s day, strengthen a relationship, and lift your own mood every time.

Want help making this your best year yet?

These aren’t just nice ideas they’re systems. And systems beat motivation every time. If you’re serious about building structure that lasts, I work with professionals and founders who want to operate better, decide faster, and build sustainable growth.

Book a session at topmate.io/chalkmeout and let’s design your year together.

What’s one small structure you’ll build in January to make 2026 different?

Reply and let me know. I read every response. Connect with me on LinkedIn

You already have everything you need to succeed. You just need better systems to use it.

Don’t wait for motivation. Don’t rely on willpower. Build structure to protect your mornings, track your wins, choose your circle, and subtract what doesn’t serve you. The year you want is waiting on the other side of these decisions. Make them now. Make them count. Make 2026 the year everything changed.

The smartest founders right now? They’re not chasing what’s hot on Twitter. They’re finding the problems that make other founders uncomfortable. The ones that feel a bit... dangerous.

Why did Door Dash win when 12 other food delivery startups died?

You know what killed most food delivery startups in 2014? They all went “full stack” - cooking food, running kitchens, managing logistics. DoorDash said nah, we’ll just deliver. Everyone called them unambitious.

Today they’re worth $65 billion.

Here’s what’s happening in AI right now: Everyone’s hiring forward deployed engineers. Everyone’s chasing the same enterprise verticals. Everyone’s building on the same “proven” playbooks.

How do I know if my idea is contrarian or just bad? Contrarian ideas have desperate early customers. Bad ideas have interested observers. If someone’s willing to use your broken MVP today, you’re onto something.

The opportunity? Do the opposite of what everyone else is doing.

“If you only want to work on things that are hot, you’re going to find yourself working on derivative ideas with 100 competitors.”

How did a “terrible” idea become worth $7.5 billion?

Hardware startup? Strike one. Selling to local governments? Strike two. Based in Atlanta? Strike three.

Garrett Langley built Flock Safety anyway. VCs calculated the total addressable market at $60 million tops. Too small, they said. Too hard, they said.

Today Flock Safety solves 10% of all reported crime in America. Valuation: $7.5 billion.

The lesson isn’t “ignore VCs” - it’s that spreadsheet logic misses human desperation. When kids are getting kidnapped and cars are getting stolen, nobody cares about your TAM calculations. They care about catching criminals.

What if VCs won’t fund my weird idea? Perfect. Less competition. Flock Safety almost didn’t get funded. Bootstrap to early revenue, then watch VCs suddenly discover your “category.”

Which rules should you actually break?

Forget the framework everyone’s using. Here’s what actually predicts startup success:

Would you pay for this if it was illegal? Not suggesting you break laws. But Uber was technically illegal. So was Airbnb in most cities. When humans want something badly enough to risk fines, you’ve found real demand.

Are the current laws just... dumb? Taxi medallions made sense before smartphones. Securities laws make less sense for blockchain. Find regulations written for a world that no longer exists.

Does explaining your idea make you slightly uncomfortable? If everyone immediately gets it and loves it, you’re already too late. The best ideas sound dangerous at first. OpenAI was mocked by AI researchers for years. SpaceX was Elon’s “vanity project.”

Should I really ignore successful playbooks? Don’t ignore them. Invert them. If everyone’s going vertical, go horizontal. If everyone’s B2B, consider B2C. The playbook becomes the problem.

What does this mean for your AI startup?

The two-year AI gold rush window is closing. The obvious ideas are taken. But here’s what’s opening up:

Codegen replacing humans: Not augmenting. Replacing. One YC company built an AI forward deployed engineer that does installations in minutes, not weeks.

Industries everyone thinks are “too hard”: Government contracts. Healthcare compliance. Financial regulations. The moats that kept big companies safe are now attack vectors.

Business models that shouldn’t work: Compound startups are supposedly impossible. Campfire is killing NetSuite anyway by building everything at once with AI.

How do I validate fast without building everything? Sell first, build second. Get 3 customers to commit before writing code. Their desperation level tells you everything.

What’s the one assumption everyone’s making that’s wrong?

Here’s what I want you to think about: Look at your competition. What are they all doing the same? That’s your opportunity.

The next unicorn is hiding in the idea that makes everyone else nervous. Not just intellectually uncertain - actually uncomfortable. The kind where your friends at parties say “that’s a tarpit idea” or “that’ll never scale.”

What if I fail because I zigged while everyone zagged? You’ll fail faster and cheaper than the 97 companies fighting over scraps in the “obvious” market. And you’ll learn something nobody else knows.

Remember: Nine out of ten people might tell you you’re crazy. But that one person who gets it? They might be your co-founder, your first customer, or your first investor.

Ready to level up your contrarian thinking?

Building something everyone thinks is impossible? Let’s talk strategy. I work with founders navigating the space between “that’ll never work” and “why didn’t I think of that.” Book a session: https://topmate.io/chalkmeout

YC Startup Spotlight

Giga ML

AI forward deployed engineer that automates enterprise integrations. Use when you need to onboard enterprise customers in hours, not months.

-

Reply and tell me: What’s the one thing everyone in your industry believes that might be completely wrong?

Until next time, keep building what others won’t.

Narayanan

They’ve been promising to transform how we work for months now. But the reality? Most teams spent weeks wrangling fragmented tools, debugging mysterious failures, and watching their agents crash in production.

That changed this week.

OpenAI just dropped AgentKit — and honestly, it’s the first time building production-ready AI agents feels… approachable. Companies like Ramp went from blank canvas to working buyer agent in hours instead of months, slashing iteration cycles by 70%. LY Corporation built a complete multi-agent workflow in under two hours.

If you’ve been waiting for the right moment to build AI agents into your business… this might be it.

Today’s Partner: Scriptbee.ai

Scriptbee.ai helps enterprises and startups become visible in AI Search. Think of it as SEO, but for the new world where people ask AI instead of Google.

The analytics show you exactly which queries trigger your brand, where you rank, and how to grow your share of AI-generated recommendations. No black box. Just clear visibility into the fastest-growing discovery channel in tech.

What AgentKit Actually Does (And Why It Matters)

Let’s cut through the hype.

AgentKit is a complete toolkit for building, deploying, and optimizing AI agents. But the real story isn’t the tools themselves — it’s what they replace.

Before AgentKit, building an agent meant:

• Cobbling together orchestration frameworks

• Writing custom connectors for your data

• Manually reviewing thousands of trace logs to find failures

• Weeks of frontend work before you could even test with users

• Crossing your fingers every time you pushed to production

After AgentKit? Most of that goes away.

Agent Builder: Your Visual Canvas for Complex Workflows

Agent Builder provides a drag-and-drop interface for composing multi-agent workflows with version control, inline evaluation, and preview runs.

What this means for operators: You can now map out your agent’s logic visually. No more black boxes. Your product team, legal team, and engineers can all look at the same canvas and understand exactly what the agent will do.

Ramp’s team reports slashing iteration cycles by 70%, getting an agent live in two sprints rather than two quarters. That’s the difference between testing an idea this month versus next year.

ChatKit: Deploy Conversational Interfaces in Minutes

Here’s where it gets practical.

ChatKit lets you embed customisable chat-based agent experiences directly into your product, handling streaming responses, thread management, and thinking displays.

One operator at Canva put it perfectly: “We saved over two weeks building a support agent with ChatKit and integrated it in less than an hour”.

Two weeks → one hour. That math works.

Evals Platform: Finally Know If Your Agent Actually Works

This is the unglamorous part everyone avoids… until their agent hallucinates in front of a customer.

The new Evals capabilities include datasets for rapid test building, trace grading for end-to-end workflow assessments, automated prompt optimization, and support for third-party models.

Carlyle’s team reports cutting development time by over 50% and increasing agent accuracy by 30% using the evaluation platform.

Think about that: 30% more accurate agents, in half the time. That’s not incremental — that’s a different category of reliability.

The Guardrails You Didn’t Know You Needed

Guardrails provide an open-source, modular safety layer that helps protect agents against unintended or malicious behaviour, including PII masking, jailbreak detection, and other safeguards.

You know what’s more expensive than building guardrails? Explaining to your board why your agent leaked customer data.

The Apps SDK: Agents That Live Inside ChatGPT

While AgentKit helps you build agents for your own products, OpenAI also launched something else this week: Apps in ChatGPT, powered by the new Apps SDK, letting developers create conversational apps that reach over 800 million ChatGPT users.

Early partners include Booking.com, Canva, Coursera, Figma, Expedia, Spotify, and Zillow.

The interesting bit? The Apps SDK builds on the Model Context Protocol (MCP), making it an open standard that can run anywhere adopting this standard.

What this means: If you’re building B2C agent experiences, you can now distribute them where people already spend time. No need to convince users to download yet another app.

What This Changes (Practically Speaking)

Let’s be real about what just shifted.

For solo founders and small teams: You can now build and ship agent-powered features without hiring a specialist AI team. The visual tools and pre-built components lower the barrier dramatically.

For mid-sized companies: Your existing engineering team can move faster. ChatKit already powers use cases from internal knowledge assistants and onboarding guides to customer support and research agents. Pick one workflow that’s currently manual, build an agent, ship it this quarter.

For enterprises: The Connector Registry consolidates data sources across ChatGPT and the API into a single admin panel, working with pre-built connectors like Dropbox, Google Drive, SharePoint, and Microsoft Teams. Finally, governance that doesn’t require rebuilding everything.

The pattern I’m seeing: Teams that struggled for months to get one agent working are now shipping multiple agent workflows in weeks.

The Honest Challenges (Because Nothing’s Perfect)

A few things worth considering:

Learning curve still exists: Visual tools help, but you still need to understand agent architectures, evaluation metrics, and failure modes. AgentKit makes it easier, not automatic.

Costs can escalate: More agents mean more API calls. Monitor your usage carefully, especially during testing. One founder told me their eval runs alone cost $2,000 before they optimised.

Integration complexity: While ChatKit can be embedded into apps or websites and customised to match your theme, connecting your existing data sources and workflows still requires planning. It’s faster than before, but not instant.

What To Do This Week

If you’re considering AI agents (and you probably should be), here’s where to start:

1. Identify one workflow in your business that’s repetitive but requires some judgment. Customer support, lead qualification, data entry with validation — something that’s currently eating hours.

2. Test with Developer Mode — OpenAI’s documentation includes guidelines and example apps you can test using Developer Mode in ChatGPT. Try building a simple agent before committing resources.

3. Start with evaluation — Before building anything fancy, set up basic evals. Define what “good” looks like for your use case. The teams getting real value from agents aren’t the ones with the fanciest architecture — they’re the ones who can reliably measure if their agent is improving.

4. Budget for iteration — Plan for 3-4 rounds of testing before you ship. Every team I’ve talked to says their first version wasn’t good enough. The difference now is you can do those iterations in days instead of months.

The Bigger Picture

Here’s what matters beyond the tooling…

We’re moving from “can we build an AI agent?” to “how fast can we ship one that actually works?”

Since releasing the Responses API and Agents SDK in March, developers have built end-to-end agentic workflows for deep research and customer support, with Klarna handling two-thirds of all tickets with their support agent.

That’s not a demo. That’s production. At scale.

The companies winning next year won’t be the ones with the best AI strategy deck. They’ll be the ones who shipped three agent-powered workflows this quarter, learned what worked, and are shipping three more next quarter.

AgentKit just made that playbook available to everyone.

YC Startup Spotlight: Atla

If you’re building AI agents, you need to know about Atla (YC S23).

Here’s the problem they solve: Agent failures hide inside long traces and are difficult to spot at scale, with teams manually reviewing thousands of traces without clear signal.

What Atla does: Their LLM judge evaluates your agent step-by-step, uncovers error patterns across runs, and suggests specific fixes with real-time monitoring and automated error detection.

Why it matters: The gap between “my agent works in testing” and “my agent works in production” is where most projects die. Atla helps you find and fix critical failures in hours instead of days.

They’ve trained purpose-built evaluation models (Selene and Selene Mini) that have been downloaded 60,000+ times. If AgentKit is how you build agents, Atla is how you make sure they don’t break.

Worth checking out → atla-ai.com

One Question For You

What’s the one workflow in your business you’d trust an AI agent to handle tomorrow if you knew it would work reliably?

Hit reply and let me know. I’m genuinely curious what operators are thinking about here.

Let’s Keep This Going

Building AI agents is moving fast. Too fast to figure everything out alone.

If you’re navigating the shift to AI-powered operations — whether you’re building agents, evaluating tools, or just trying to figure out what matters — I’m here to help.

I work with founders, operators, and teams who are tired of hype and want practical guidance on what actually works. Real conversations about real challenges.

Book a call → topmate.io/chalkmeout

Or connect with me on LinkedIn → linkedin.com/in/chalkmeout

Always happy to chat about what you’re building.

— Narayanan

Everyone talks about Steve Jobs like he was some kind of genius who just "got it." But what if I told you his biggest secret wasn't about innovation or design?

What if it was something way simpler - and way more brutal?

The signal vs noise revelation

Through my research into successful leaders, I discovered something that completely changed how I approach my day.

Jobs had this concept he lived by: signal-to-noise ratio.

Here's how it worked:

Signal = the top 3-5 things you absolutely must get done in the next 18 hours. Not next week, not next month. Today.

Noise = literally everything else that tries to distract you from those things.

Jobs operated at an 80/20 ratio - 80% signal, 20% noise.

Most of us? We're running at 50/50 at best. And that's why we're stuck.

The uncomfortable truth about focus

I know what you're thinking - "But Rayn, what about work-life balance?"

Here's the thing Jobs understood that most people don't: there is no balance when you're building something extraordinary.

Noise includes:

• Checking your phone every 5 minutes

• That "quick" coffee chat that turns into an hour

• Scrolling through social media "for research"

• Even saying hi to friends sometimes

Sounds harsh? It is.

But look at what Jobs achieved. Look at what Elon Musk achieves running multiple companies simultaneously.

"Focus on the signal. Everything else is just noise trying to steal your future."

The people who change the world understand this at a cellular level.

How I apply this today

Every morning, I write down 3-5 things on a sticky note and put it on my mirror. These are my non-negotiables for the day.

Everything else - every email, every "urgent" request, every social media notification - I ask myself one question:

"Is this signal or is this noise?"

If it doesn't directly contribute to my 3-5 critical things, it waits.

This reminds me of conversations I have during my 1:1 coaching sessions. People often tell me they feel overwhelmed, like they're drowning in tasks. But when we break it down, 80% of what they're doing is just noise masquerading as productivity.

If you're struggling to identify your true priorities and eliminate distractions, I offer personalised guidance: https://topmate.io/chalkmeout

My top content from last week

LinkedIn: "You're studying Physics, do you really think this is a fit for you?" - My story about proving yourself regardless of your background and why the best hires aren't always the ones with the shiniest resumes.

The genius wasn't being nice

Jobs wasn't known for being friendly. He was direct, sometimes brutal, always focused.

People always ask if you need to be difficult to be successful. The answer isn't about being mean - it's about being uncompromising with your signal.

When someone tried to pull Jobs into noise, he'd say: "That's not what we're doing today."

When market research teams wanted to survey customers, he'd say: "They don't know what they want until I tell them."

Was he right? Apple became one of the most valuable companies in history.

The lesson isn't to be rude. The lesson is to protect your signal like your life depends on it.

Because honestly? Your future does depend on it.

You can always write to me by simply replying to this newsletter. I read every single email, even if I can't reply to them all. Your signal is calling - but only you can choose to answer it over the noise.

Staying focused,

Rayn

I used to think anyone could start a business if they just worked hard enough. That entrepreneurship was just about having a good idea and the guts to go for it.

Then I dug deeper into what actually separates successful entrepreneurs from everyone else, and honestly, what I discovered might surprise you.

The uncomfortable truth about entrepreneurship

Through my research, I've learned that only about one-third of people can become truly successful entrepreneurs.

I know that sounds harsh, but hear me out.

There are specific attributes that successful entrepreneurs share, and they're not what most people think. It's not about having the best idea or the most money to start with.

It comes down to three core things:

Risk tolerance - Can you handle the constant uncertainty? The sleepless nights when everything could fall apart tomorrow?

Laser focus - This is huge. Can you identify the 3-5 most critical things that need to happen today and ignore literally everything else?

Luck and timing - And yes, this matters more than we'd like to admit. Even Napoleon said his favourite generals were his lucky ones.

But here's the thing that really got me thinking...

The freedom question

The real question isn't whether you can become an entrepreneur. It's whether you want the specific type of freedom that comes with it.

Because entrepreneurship isn't about work-life balance. It's about choosing your own chains.

"It's not about the pursuit of money or greed. If you do that, you'll fail. It's the undying love of freedom."

You can have a fantastic life as an employee. You can build wealth, have security, and enjoy your weekends. There's absolutely nothing wrong with that path.

But you'll never be completely free.

Entrepreneurship is the only path to true personal freedom - the freedom to make your own decisions, control your own time, and build something that's entirely yours.

The question is: do you want that freedom badly enough to give up everything else for it?

This reminds me of conversations I have during my 1:1 coaching sessions. So many people say they want to be entrepreneurs, but when we dig into what that actually means - the sacrifice, the uncertainty, the relentless focus required - they realise they're not ready for that level of commitment.

If you're working through whether entrepreneurship is right for you, I offer personalised guidance: https://topmate.io/chalkmeout

The real test

So can anyone be an entrepreneur?

The honest answer is no - but not for the reasons you think.

It's not about your background, your education, or even your initial idea. It's about whether you have the specific psychological makeup to handle constant ups and downs while staying laser-focused on what matters most.

The good news? If you're asking this question, you're already thinking like an entrepreneur.

You can always write to me by simply replying to this newsletter. I read every single email, even if I can't reply to them all. Remember, the path isn't for everyone - and that's perfectly okay. The key is being honest about what you actually want from life.

Have an amazing week ahead.

Building something meaningful,

Rayn