AI Automation Engineer (self-taught) · n8n · LLM agents · production systems I turn language models into systems that actually run — and keep running.
I build AI-powered automation for production, not demos. Self-taught, I work the way engineering actually works in 2026: integrating LLMs, wiring data pipelines, designing agent loops, deploying, and keeping systems reliable once they're live.
My approach: chase root causes, verify the live state instead of trusting assumptions, and build the tooling that measures truth rather than confirming hope.
🤖 Multi-agent trading-analysis engine — end-to-end on n8n
Live market data → a deterministic market-structure (SMC/ICT) core → three LLM analysts
(fundamental / technical / structure) → an adversarial verifier → a deterministic
backtesting & validation harness. Heavy on validation discipline: walk-forward,
out-of-sample, expectancy in R, Monte-Carlo drawdown — and a single deterministic source
of truth that the LLM layers read rather than re-derive.
n8n · Claude · Gemini · OpenRouter · Python · REST/webhooks
🧠 Personal knowledge-automation OS
A "second brain" wired to run itself: n8n cloud↔local bridges, scheduled brief/review
agents, an inbox pipeline, and health-monitoring that checks its own liveness.
n8n · Obsidian · MCP · cron · Telegram
🛡️ Security & reliability tooling
Secret-leak guardrails (pre-commit gates + content scanning), install-time skill vetting,
encrypted private backups, and a disaster-recovery runbook.
gitleaks · pre-commit hooks · Python
Automation: n8n · Webhooks · Cron · Data pipelines AI: Anthropic Claude · Gemini · OpenRouter · MCP · Claude Code · Prompt & context engineering · Agent loops · RAG Engineering: Python · Git · REST APIs · JSON · Debugging · Root-cause analysis · System design Quant: Backtesting · Expectancy · Monte-Carlo · Walk-forward validation
- Building & validating production AI systems — in public.
- Open to remote roles & freelance in AI automation / LLM engineering.
📫 LinkedIn · ✉️ (13tomenko@gmail.com)