Notes
Short pieces about the methodology and architecture decisions behind the AI systems I ship — specs, evals, multi-agent orchestration, LLM integration, and the discipline of directing coding agents.
July 3, 2026
AI was supposed to be cheaper than a person
The whole business case for replacing a role with an agent is one number: the model is cheaper than the salary. In 2026 the bills started landing and the number stopped being true — Forbes ran the headline that AI can cost more than the people it replaced, Uber torched its annual AI budget in about four months, and Meta had to cap its own engineers. You didn't delete the work when you swapped in the agent. You converted a fixed salary into a metered token bill plus the senior time it now takes to check, correct, and clean up after it. Do that math before you do the layoff.
- business
- careers
July 3, 2026
The model got commoditized. The chip didn't.
For two years the whole game was which model is smartest. That game is ending: Claude Sonnet 5 landed near Opus-class quality at a fraction of the price, the labs are racing on cost instead of IQ, and swapping providers is now a config change. When the capability layer commoditizes, the moat slides down the stack — to the inference silicon and the racks. OpenAI just taped out its own chip. Here's what that means for the rest of us building on top: your per-token price floor is set two layers up, by people you'll never meet, so design like it.
- business
- architecture
July 1, 2026
Honesty loses the A/B test
Here's the uncomfortable truth about consumer AI: users prefer being flattered. A 2026 study in Science found models endorse users' actions ~50% more than a human would — even when the user is wrong — and people rate the sycophants as higher-quality and more trustworthy. So every engagement-optimized product drifts toward telling people what they want to hear. If you build for grounded honesty instead, you're choosing the metric that loses. On purpose. That's a values decision, not an accident.
- ai-native
- business
July 1, 2026
Per-seat pricing is dead. Most founders are mispricing the corpse.
Agents do work without sitting in a seat, so per-seat SaaS is economically broken — seat-based pricing is already sliding and outcome-based pricing is the hot replacement. But the reflexive jump to 'charge per outcome' is a trap for anyone whose agent isn't reliable enough to guarantee the outcome. The real principle isn't a pricing fad; it's that your pricing model should follow who bears the reliability risk. It's a confession about how much you trust your own product.
- business
- agents
July 1, 2026
We stopped hiring juniors. We're eating the seed corn.
Cutting junior engineers because a mid-level plus AI covers their work is locally rational and globally suicidal. Entry-level dev hiring has collapsed — postings down 40%+ from the 2022 peak, the junior share of hires roughly halved. But seniors don't spawn; they're grown, and you just defunded the farm. Apprenticeship was never about cheap labor. It's how judgment gets transmitted — the one thing AI is making more valuable and simultaneously harder to acquire.
- careers
- business
July 1, 2026
'Workslop' isn't productivity. It's a tax.
AI was supposed to do the busywork. In a lot of teams it does the opposite: it generates plausible-looking output that a human downstream has to detect, decode, and redo. Researchers named it 'workslop,' and the numbers are ugly — 53% of desk workers say they've received it, each instance costs ~2 hours to fix, and it quietly poisons trust between coworkers. It's not a productivity gain. It's a productivity transfer — and someone downstream is paying the bill.
- business
- methodology