Jonny

More ideas than time. Every post is mine — no AI ghostwriter.

[Jonny @jonny_hein]

Your new AI receptionist will destroy your company.

Sure the cost savings and efficiency are easy to see. What's less obvious is what you lose: the personal service people crave in an era of bots.

As everything gets automated, people become the premium worth paying for.

[The All-In Podcast @theallinpod]

Chamath: The AI Boom Has a Borrowing Problem

“The entire United States economy is levered to the AI complex.

So very narrowly, we're talking about seven or eight companies that set the tone for the entire country. And whether we like it or not, that is where we are.

Why do I say that? The most important companies in the world are in a massive spending binge, right?

These were highly profitable, highly cash accretive, highly cash flow generative businesses that have said, ‘We're going to put the foot down on spending because we believe in what's coming into the future.’ That's what drives GDP and it's what drives growth.

Now, as they do that though, they deplete their cash and they have to raise debt.

The AI complex now owes $340-400 billion to the market. And you'd say, ‘Okay, why is that important?’

The problem is this: These companies are raising so much money it starts to crowd out the capital markets.

And the most important part of the capital market that it starts to crowd out, is the appetite for government treasuries.

And what you can see here is that when you look at these AI firms, the amount of money that they need to raise through the end of '27, it's 30% of all of what the Treasury's about to issue in that same period.

So, whether we like it or not, they are going to start competing for demand.”

[Jonny @jonny_hein]

AI telemetry at work: useful insight or employee surveillance?

There’s a lot a company could learn from how people use AI. Where’s the line between improving DX and invading privacy?

[Jonny @jonny_hein]

@github just announced that agents start more than 40% of pull requests on GitHub, up from 17% in March. That's 110M in September alone.

They're merging fast too. 90% of agent-involved PRs merged within two weeks in September, versus about 86% for PRs with no agent detected.

But GitHub's Octoverse report stops at merge. It has nothing on reverts, incidents or rework, which is the part your team actually lives with.

If you're rolling out agents, tag every change agent or human, then track what happens after merge. That's how you find out whether fast also means good.

Which post-merge signal would convince you agent PRs are working? 👇

Full report (GitHub measuring its own platform): https://t.co/YyQ1n63CJc

[Jonny @jonny_hein]

There is a market for more software security auditing companies

Let the business owners, and non-technicals vibe-code the solution, then pay you for the peace of mind

[Jonny @jonny_hein]

Getting a vibe-coded solution working? Easy. Knowing it’s secure and reliable? Still hard.

Especially when you treat your entire application as a black box system.

[Jonny @jonny_hein]

@levelsio All the bot accounts deciding if they should reply

[Jonny @jonny_hein]

@PeterDiamandis Comparing AI to religion is crazy.. maybe hop off X for a bit

[Jonny @jonny_hein]

@trq212 The world is full of geniuses waiting for permission

[Jonny @jonny_hein]

@jasonlk $1B ARR and the ability to collect and monetize hard-to-collect data

[Jonny @jonny_hein]

2020. Live money on Binance.

My first real project: a Bitcoin bot scraping 3–10¢ profit per trade and recycling it into buying power.

Sheet below. 2/20 groups. ~$325 each.

They banned trading in my state shortly after. Didn’t make more than a couple hundred bucks.

Still the cleanest automation lesson I’ve learned: tiny edges + feedback loops beat big opinions.

[Jonny @jonny_hein]

@SyntheticBeef Ye ye ye all for making these saas companies work for it. Tell me what I want to hear: you have a data company not a saas company

[Jonny @jonny_hein]

@felixrieseberg DX is more important than ever

Show me you understand the impact of empowering developers, removing friction, and creating an environment where engineers can do their best work, and you’re hired

It’s that simple

[Jonny @jonny_hein]

Been grinding to get AI thought-slop out of my FYP

Brick by brick baby 🧱

[Jonny @jonny_hein]

@tech__unicorn Is Silicon Valley still a place people think smart people live?

[Jonny @jonny_hein]

@Kedeodev Unless your stance on ai consciousness is similar to @DarioAmodei then you have a startup with 5 employees that demand model welfare

[Jonny @jonny_hein]

@Doudoujay 100%

[Jonny @jonny_hein]

The credits hurt but the real bill is attention. Every time the agent retries the same fix 6 times you’re paying with judgment time reviewing bad output. That’s the expensive part… I do think the frontier models are getting a lot better at knowing when and what rabbit holes to explore

[Jonny @jonny_hein]

@cem_hasoglu Agents multiply whoever you already are. without those 3 exceptional people you’re just a founder with a lotttt of confident wrong output

[Jonny @jonny_hein]

Give away agents and charge for secure terraform code

The frontier models can create the crm, the dashboard, the app, and the agents, but businesses don’t want their infra to be a vibe-coded black box featured in next weeks breach roundup. They’ll pay for that peace of mind (especially when handling PII)

[Jonny @jonny_hein]

100% however there is some beauty in uncovering things along the way that you never thought to ask when you started the research. AI is good at confidently deciding what’s worth your time skipping from A directly to Z. But *sometimes* there’s value in the letters in between. Spend an hour researching cat names to read an article on why dogs are a superior pet worth adopting instead. Terrible example but my point stands 😂