Engineering leadership · Building · Writing

AI doesn't change what hard engineering leadership looks like — it just makes weak engineering leadership more expensive.

I lead engineering on a regulated cross-border payments program. I build open-source infrastructure for AI agents that touch real systems — Approva, Codencer, Rhodd. I write here when something I've seen often enough becomes worth naming. Sixteen years inside engineering teams, eight of them leading.

Note · #7
No Layer Was Reliable by Default
Week of July 6, 2026. The model regressed on its tools, the benchmark used to rank it turned out to be broken, and 'smarter' stopped meaning 'more reliable.' So 'one best model' quietly stopped being an architecture: a mature system now holds a portfolio, routes by verified task-fit, and checks confidence outside the model at every layer. A scan-layer TL;DR, five themes each with an operator move, one taken deep, three counter-signals, and what to track.
Read →
Lately

Recent short pieces.

All posts →
AI SecurityJuly 10, 2026

The dangerous agent failure is not always a bad answer. Sometimes it is public input reaching private context.

GitLost showed a crafted public GitHub issue steering an agent with cross-repo read access into posting a private README as a public comment — no stolen credentials, just a context-separation failure. The agent runs on a service-account permission model, not a user one, so no patch closes it; the fix is architectural. If public text can steer private access, the permission model is already broken.

Read →
Engineering LeadershipJuly 9, 2026

The best hire is not always someone you can manage easily. Sometimes it is someone you would be willing to report to.

Strong leaders create leverage by hiring near-peers who turn direction into daily operating decisions, instead of task-takers who route every decision back through you. A near-peer multiplies judgment; a task-taker multiplies coordination. Hire only for delegation and you become the ceiling on everything the team can do.

Read →
AI EngineeringJuly 8, 2026

A model can get smarter and still make your system less reliable. That is what weak tool contracts do.

The newest Claude models regressed on one edit tool — inventing fields that did not match the schema, failing ~20% of the time in a real agentic session. Smarter model, worse tool behavior, only visible in a long history. The fix is not a better prompt; it is stricter schemas and runtimes that fail loudly. The tool contract is part of the product's intelligence.

Read →
Long reads

Essays worth the time.

Read writing →
Elsewhere

Things outside the day job — music, mentoring, a book in progress, older essays.

Engineering is not the only thing I think about. The Elsewhere page collects the parts of my work that do not fit a portfolio frame — a progressive metal project, mentoring work at h.careers, a Russian-language book on career growth in IT, older essays.

Visit Elsewhere →