Twenty years of building software and companies, now focused on AI systems — LLM-powered apps, RAG pipelines, multi-agent architectures, and the engineering substrate around them. I take on real business problems, as an architect and as a senior developer — often both. What I care about is the result: a solution that does what your users actually need and can keep growing, not just code that runs once. If it's a worthwhile, interesting project where good engineering is what makes the difference, I want to hear about it.
Why architecture is worth paying for
The clients I value most are the ones who understand that architecture matters — who are willing to invest in the few decisions that shape everything downstream. That investment pays off. The opposite — "let's just hack something together" — usually ends in a dead end: the app doesn't do what users really need, and it can't be extended, because the code never followed the basics (SOLID, DRY, KISS, dependency injection). A handful of good decisions up front is what saves you from a rewrite a year later. I'm happy to write code — I'm a senior developer and always will be — but I want the thing being built to be worth building well.
How to engage me
Design and build, embedded
Three to six months, integrated with one team. Designing the AI system — retrieval, tool use, orchestration, evaluations, MCP integrations, data layer, security boundaries — and building it, or reviewing the build as it ships. Single LLM in the loop or a full multi-agent topology, whichever the problem actually wants. I write the load-bearing parts myself and direct coding agents for the rest; either way the spec, the design, and the review are mine. Best fit when you're past the prototype and the next version needs to survive contact with production.
Architecture review
One to three weeks, written deliverable. A senior pair of eyes on a system you've already designed, or on one that isn't behaving the way you hoped. Output is a written report: what's load-bearing, what's brittle, what I'd change first, and why. Short engagement, no ongoing relationship implied. Not sure you need one yet? The AI Architecture Scorecard is the free, self-serve version — six dimensions, a prioritized read on where you stand, in about three minutes.
Full-time
Senior AI engineering and architecture roles at companies that take engineering quality seriously and put "ship it" as the second sentence, not the first. Whether the title says architect or senior developer matters less than the work being real. Open in principle, selective in practice.
How I work
- The spec is the artifact, not the prompt and not the code. A clear specification beats a clever implementation, and it's what the team actually reasons about.
- Evals or it didn't ship. Behavior worth shipping is behavior worth pinning down — I expect a held-out scenario suite for any agent we plan to put in front of real work. My take on why.
- One source of truth per fact. Duplicated state is duplicated bugs. I look for it actively in reviews.
- Boring tech where it can be boring. Postgres, FastAPI, Next.js. Novelty belongs in the AI layer; the substrate should be predictable.
- Async by default. Long writing beats short meetings. Decisions land in documents so the next person doesn't have to rediscover them.
These principles have a fuller home in the AI-native architecture manifesto.
What I don't take on
I'm open to interesting work, but not to everything. Where I'm a poor fit:
- Pure staff augmentation — a seat to fill against a fixed spec, with no say in how the thing is built. I'm at my best when the design is part of the job, not handed to me as settled.
- Engagements with no production target. I'm interested in the systems that have to survive a Monday morning, not the demos that have to survive a Tuesday afternoon.
- Projects where "fast and disposable" is the actual goal. If the plan is to hack it together and never maintain it, I'm the wrong person — that's not where I add value.
Background
Twenty years across backend systems, distributed data, payments, LLM-powered SaaS, and now production AI architecture. Shipped at small startups and at scale; worked both on teams and as an individual. The throughline is engineering judgment under real-world constraints — the full résumé is here.
Get in touch
The fastest path is email: fedorstartup@gmail.com. Tell me what you're building, what stage you're at, and what specifically you want a second pair of eyes on. I read everything and reply within a few days; long-form is welcome. The contact form works too.