16 years of backend in production. I build LLM pipelines and AI integrations on top of Django/FastAPI — from architecture to deployment.
Backend that handles load. LLM functionality shipped to production.
LLM pipelines in production, agentic systems, structured output, integrating models (OpenAI/Anthropic API and others) into existing backends, token cost control
Django, FastAPI, asyncio, Celery, PostgreSQL/MariaDB, Redis, Docker, CI/CD — from data schema to deployment and monitoring
REST, WebSocket, payments, CRM, telephony, bots, parsers, ETL — connecting services and automating routine
MVP from idea to deploy, iterating on feedback, own SaaS products
Automated LLM pipeline: brief → script → media generation → cut, one loop without manual intervention. An LLM (Claude) writes the scene-by-scene script with a hook and voice direction; a Python/asyncio pipeline generates video, voice, music and subtitles via fal.ai, Replicate and ElevenLabs and stitches a finished vertical video up to 60 seconds with ffmpeg.
Educational platform for IT professionals with quizzes, mock interviews, code review, and personalized recommendations based on a knowledge decay algorithm
An LLM pipeline generating multi-layered personal reports from several data sources: five independent computed layers are synthesized by the model into a single structured portrait. FastAPI + PostgreSQL + Next.js, free basic result and a paid full report.
Analytics platform for university management with tools for assessing student learning outcomes and recommendations for optimizing educational processes
Cal.com for the Russian market: from architecture to production. FastAPI + React, FSD, OAuth, calendars, email, 196 tests.
Unified platform to auto‑generate conference landing pages and collect leads. Solved the problem of creating dozens of near‑identical pages for PVS‑Studio event participation.
When synchronous calls strangle your system, how to implement events painlessly, Kafka vs RabbitMQ in practice, and why eventual consistency is a conscious choice. Typical production cases with metrics and a breakdown of all EDA pitfalls.
Deep dive into ty — the new type checker from creators of uv and Ruff. Intersection types, intelligent narrowing, 10-100x faster than mypy. Revolution or just hype?
Code Review in 80% of teams is theater. 'LGTM' in 30 seconds, a checkbox before merge, wasted time. I've watched for 12 years how reviews kill productivity instead of boosting it. Here's a system that transforms CR from bureaucracy into your team's primary growth tool.
A career planning service for IT professionals — see where to grow and what to learn next.
Open to hire and contracts: Python backend, LLM pipelines, AI integrations. Describe your task — I'll reply within 24 hours