Pawel Brodzinski

Overwork is a New Normal “Anything Is Possible” has become “Everything Is Exhausting”

“We are instituting a 30 hour work week.” ~ Tech Bros in 2020
“Why can’t you stay all weekend?” ~ Tech Bros in 2025
~ @ourfounder.bsky.social

humanework.substack.com/p/the-6x9-ec...

The value of what we're delivering (as the IT industry) doesn't justify the salaries we've been getting. And I'm not talking only about the post-COVID heyday, when the disconnection went to absurd levels.

It's been that way for decades. When I joined the industry in 2000, it was already so.

The Most Underestimated Factor in Estimation - Pawel Brodzinski on Leadership in Technology Communication quality is a hidden factor that can make product development longer. Yet, it is rarely considered when estimating.

In the whole discussion about "how to estimate software," there's one aspect that's curiously almost always absent.

Quality of communication.

And yet, bad communication can easily increase the effort/cost by 100%. Or cut it by one-fourth.

brodzinski.com/2025/08/most...

As developers get to use more AI they trust it less.

A surprise? If you're paying attention, then probably not so much.

What’s a technology that you think is overhyped? I’m going to give a sideways answer to this, which is that the venture capital business model needs to be understood as requiring hype. You can go back to the Netscape IPO, and that was the proof point that made venture capital the financial lifeblood of the tech industry. Venture capital looks at valuations and growth, not necessarily at profit or revenue. So you don’t actually have to invest in technology that works, or that even makes a profit, you simply have to have a narrative that is compelling enough to float those valuations. So you see this repetitive and exhausting hype cycle as a feature in this industry. A couple of years ago, you would have been asking me about the metaverse, then last year, you would have asked me about Web3 and crypto, and for each of these inflection points there’s an Andreessen Horowitz manifesto. It’s not simply that one piece of technology is overhyped, it’s that hype is a necessary ingredient of the current business ecosystem of the tech industry. We should examine how often the financial incentive for hype is rewarded without any real social returns, without any meaningful progress in technology, without these tools and services and worlds ever actually manifesting. That’s key to understanding the growing chasm between the narrative of techno-optimists and the reality of our tech-encumbered world.

Stand by this: www.politico.com/newsletters/...

Journey Hypothesis or Destination Hypothesis? It's about a journey, not a destination. Building a startup is exploration and not following a pre-planned route. Even if we make such stories in retrospect.

So many products are built as if we knew the destination. Then it's easy to design a route and follow it.

Save for some edge cases, it's a wrong metaphor. We don't know what we'll build eventually.

It's unknown. What's more, it's unknowable up front.

pawelbrodzinski.substack.com/p/journey-hy...

Y Combinator CEO to the founders: "You don't need 100 engineers."

A little-known secret: You've never needed them. It's just what VCs pushed down your throat for decades.

What's new again?

I argued for decades against fixed-price fixed-scope deals with the argument that the scope is never fixed.

It can be (and will be) interpreted differently by both parties.

While we both understand $100k the same way, "setting administrative rights to users" leaves a lot of wiggle room.

The only map describing the terrain precisely must be of the size of the terrain.

The only specs defining the code precisely must be of the size of the code itself.

Having said that, though:
- Maps/specs are models.
- All models are wrong. Some are useful.

In product development, we typically start with the destination hypothesis: The destination is known; we can devise a route and follow it.

In reality, we should go with the journey hypothesis: The destination is unknown and unknowable; we need to focus on exploration.

I deem that tweet genuine, i.e. not generated by AI.

You get enough of those and it will go through the filters of a broader audience.

It's gonna be the new "like"

;)

My take on AI-generated content:

Imagine we have 10x as many new posts, articles, etc. Or better, 100x as many.

The bad news is that we still have only 1x as much attention (probably less as our attention span gets markedly shorter).

How does such an endgame look?

"The first rule of Dunning-Kruger Club is that you don't know you're in Dunning-Kruger Club." ~ @derekwwade.bsky.social

BTW: as much with offshoring as with AI a huge part of new cost incurred is in communication (or lack thereof).

The code output doesn't matter if you're building a wrong thing.
Or you take on huge tech debt without consideration or conscious decision.

My frame for that:

Imagine that you could get a team of 10 devs for free.
They'll work to the best of their abilities.
They'll never ask any question.

They. Will. Never. Ask. A. Single. Goddamn. Question.

Would you hire such a team?

Bootstrapping is a tough path.

You can't go extravagant with building early versions of a product.
You need to optimize for business sustainability, not fast growth.

But all these things, they prime founders to make better product decisions.

Received a salary report summary (for IT):

* IT salary satisfaction is declining
* Pay expectations are up 27% on average for job changers
* 85% would quit if they lost remote work options

Let me rephrase: the job market's getting worse, so we increase our expectations. Go Attaboys!

Oh, I am totally 10x more productive in terms of code output.

Without AI, I've just been making ocassional programs whenever spreadsheets wouldn't be enough.

Now, I generated 2 app with Lovable. I bet it's 10x more code (haven't looked, though, was too scared what I'd see) ;)

Mira Murati's Thinking Machines Lab is worth $12B in seed round | TechCrunch Former OpenAI CTO Mira Murati says that her new startup, Thinking Machines Lab, was valued at $12 billion in its seed round.

Thinking Machine Labs secured $2B seed funding.

Startup ecosystem cheers!

Me: Let me rephrase. A product company that *hasn't announced* what they work on (let alone validated *anything*) is worth $12B? This whole ecosystem is broken.

techcrunch.com/2025/07/15/m...

As a reference, I like this experiment:

Toss a coin 10,000 times. Add 1 for heads, detract 1 for tails. What's the most likely sum at the end?

It's 0, obviously.

But what's the likelyhood of the sum after one series to be 0? Surprisingly low.

I'd speculate the probabilities would be super low.

I imagine it like an extremely flat Gaussian distribution. There's "the most probable" output. But because of so many possible outputs, the probabilities must be low.

Not sure how we'd react to see: "I'm like 1.27% sure it's the best answer" :)

Dear Startup, Build Less The single most critical advice for founders? Build less. The single most useful leverage to achieve that? Overconstrain your budget.

If there's one bit of advice I'd give any founder, it would be this:

Build less.

Seriously. That's it.

pawelbrodzinski.substack.com/p/dear-start...

We can't trust AI to tell us when we can't trust AI.

It's funny how much this observation affects how we design atcual AI-powered features.

We consider false-positives, false-negatives, precision bands, etc.

The biggest challenge, though? Uncertainty, about when it's a guess.

I like Liz Keogh's frame: sadlines.

What happens when we miss the date?
Nobody's gonna die. Someone's gonna be sad.

This small trick removes a lot of unnecessary fears around time-sensitivity.

I always preferred "so, here's my war story," shared over a bonfire, over "thou shalt do that," spoken from an ivory tower.

Sadly, most events, newsletters, social media even, are dominated by the latter.

Turning the Standup into the Starting Block Turning wastes of time into something else.

"To be clear, huddles and standups are extremely important, most of them are just wastes of time."
~ @ourfounder.bsky.social

substack.com/inbox/post/1...

VCs: Paying Culture Lip Service Don't listen to investors' "Don't screw up a culture" advice. Their expectations of rapid growth render such a goal impossible to reach.

Exactly!

No matter what they claim, "an investor wants enough of their investments to go through the roof and doesn't care much about the rest. A lingering startup that may require another funding round is more of a liability than an opportunity."

pawelbrodzinski.substack.com/p/vcs-paying...

some of you may die written on a poster with shrek and knights ALT: some of you may die written on a poster with shrek and knights

Every single VC to their startup portfolio.

Waiting for the first startup success story like this:

“It was bleak. Our burn was $50K/mo: $25K salaries, $25K in tokens. Then we made a radical decision:

We cut our token spend to $0.

Everyone said it’s crazy. But we doubled runway, and gave us the time to do the pivot!”

The Renaissance of Full-Stack Developers - Pawel Brodzinski on Leadership in Technology Full-stack developers are back. AI demands broader skills, reversing the specialization trend we've seen for decades.

I was never particularly fond of the trend for increased specialization in software engineering.

I think we've just turned the tide. "Anything it takes" kind of devs (which is like a full version of full-stack) are back.

brodzinski.com/2025/07/rena...