AI Architect CV
Igor Tarasenko
AI architect and senior software engineer in Amsterdam. Ex-Uber. Lead iOS engineer. Practical AI systems, agentic workflows, and production mobile AI.
Download PDF → Download Markdown
Summary
16 years building production software across mobile, platform, payments, developer tooling, and applied AI. I design AI systems that can survive real codebases: evals, guardrails, MCP tools, on-device inference, cost/latency budgets, and team adoption workflows.
Current focus
- Agentic engineering workflows for iOS and product teams.
- LLM features that move from prototype to production.
- On-device AI and privacy-first mobile architecture.
- AI developer tooling: MCP, Claude Code, Cursor, repo conventions.
Experience
DNA.inc — Lead iOS Engineer & Internal AI Enablement
Lead iOS engineering while also acting as the practical AI adoption lead for engineering workflows: internal AI sessions, agentic coding patterns, context files, review loops, and guardrails.
Uber — Senior Software Engineer
Worked on platform and payments systems at Uber scale: server-driven UI, internal developer infrastructure, payment reliability, and multi-team architecture work where small error rates became real incidents.
Independent / Open Source
Built WhisperBoard, mcp-browser-use, VibeSwitch, and local ComfyUI / Stable Diffusion-family pipelines. Mentor engineers through Codementor on iOS, AI integration, debugging, and architecture.
Selected proof
- WhisperBoard: 50k+ downloads, 4.8 App Store rating, 1,035 GitHub stars.
- mcp-browser-use: MCP server for browser automation by AI agents, 938 GitHub stars.
- Agentic iOS: built preview-feedback tooling before Apple shipped native agent support.
- Codementor: practical mentoring on iOS, AI integration, and debugging.
Stack
Swift, SwiftUI, UIKit, TCA, Python, TypeScript, Anthropic/OpenAI APIs, Claude Code, Cursor, MCP, ComfyUI, Stable Diffusion-family workflows, Postgres/Supabase, FastAPI, Hetzner, Coolify, Vercel.
For the longer narrative version, see experience.