Engineer, 7 years building production systems: backends, pipelines, automation, end to end. Background in fintech and high-load (ex-Yandex, EPAM). For the last couple of years, deep in AI: RAG, agentic pipelines, model evals.
Mostly Python (FastAPI, LangGraph, SQLAlchemy), Postgres, Docker; some TypeScript/Astro.
Fixed-price Technical & Security Health Check for products and MVPs: security, scaling, code quality. See the offer →
- ready-for-traffic — launch-readiness checklist for founders shipping a product, vibe-coded apps included: 8 areas from the server to the monthly bill. CC0.
- codegen_orchestrator — multi-agent orchestrator: a Telegram brief becomes a deployed project (code, CI/CD, domain, SSL).
- service-template — spec-first Copier framework for AI-agent-ready microservices.
- personal_site — self-hostable personal-site template (Astro + FastAPI/Postgres). Powers vladmesh.dev.
- dnd-simulator — text RPG with multi-level LLM world simulation.
Privacy, civil liberties, and AI safety: the space behind Tor, Matrix, independent journalism, and open AI-safety tooling. I contribute a bit where I can, and want to do more over time.
- Site: https://vladmesh.dev
- LinkedIn: https://www.linkedin.com/in/vladmesh
- LessWrong: https://www.lesswrong.com/users/vladmesh
- Telegram: https://t.me/vladislav_meshk