AI Product Manager · AI Implementation Engineer (LLM, RAG)
I turn a business problem into a working AI service and take it to production myself —
from process analysis to deployment on the server. I use AI every day: one person plus AI delivers
the output of a small team.
Ufa, Russia · remote / hybrid · full-time, part-time or project-based ·
contacts in the CV and application
What I cover end to end: process analysis (BPMN) → solution design →
Python code → server set up and configured by me (Linux, Docker, automatic HTTPS) →
API integrations → user training → support in production.
458 hof focused work with AI (over 114 working days)
48AI products and services
508,811lines of my own code
1,186search visits to the flagship (past year)
Models I work with: Claude · GPT · DeepSeek · LLaMA · Qwen (local, Ollama) ·
vision-LLM · TTS — I pick the model by task and cost.
Volume of work: 45.3 billion tokens · 10,635 documents · 17.1 million data cells processed.
Figures come from my own portfolio dashboard (hours, tokens, lines of code, search visits from
Yandex.Webmaster and Google Search Console).
Access infrastructure for AI platforms — built a single secure perimeter giving a team
reliable access to international LLMs (Claude, GPT and others) for day-to-day work.
VPN infrastructure — own network of VPN servers (VLESS/Reality), key dispenser,
censorship bypass; part of the infrastructure block (85 h on servers, VPN, deployment, backups).
FriendlyAI (friendlyai.ru) — “AI agents without developers” and “AI + 1C” for construction
companies: showcase and rollout.
AI-Shtat — turnkey AI adoption for business: RAG consultant, implementation playbooks.