Services

What I build and fix

Four engagement types, each backed by production case numbers you can read on this site: a backend MVP shipped in 60 days, a 109M-row PostgreSQL rescue, a queue layer running thousands of concurrent AI jobs, and production AI applications.

How the engagements run

Every engagement starts with a written plan: what changes, what it costs, and the risks accepted on purpose. Then work ships in deployable weekly increments — no multi-week silence followed by a reveal.

I work as a senior engineer either embedded in your existing codebase or owning a contained subsystem end-to-end. Remote, timezone-friendly for EU and US East Coast; direct contact via email or Telegram, no account managers in between.

If your problem sits somewhere between these four pages, describe it — the first conversation is free and an hour of it usually tells both sides whether this is my kind of project.

FAQ

Do you work fixed-price or hourly?

Continuous work (a rescue, an embedded senior engineer) runs hourly with a weekly written summary. Contained deliverables — an MVP, a specific migration — are quoted fixed-price after the architecture week, because a written plan makes fixed prices honest.

How does a typical start look?

A call or Telegram conversation about the constraint that actually hurts, then a written scope with milestones within two business days. You own the plan and can throw it away if it does not convince you — that review is also the cheapest way to check the fit.