AI Product Engineer building production software with LLMs, structured memory, and intelligent workflows.
I design and ship AI-native products that combine LLMs, structured memory, workflow automation, and great developer experience.
Over the last 6 years I've delivered B2B SaaS products, led cross-functional teams, and recently focused on building AI systems that solve real customer problems — not AI demos.
I build AI systems that people actually use in production.
- AI copilots
- AI agents
- RAG systems
- evaluation platforms
- workflow automation
- TypeScript & Python applications
- APIs and integrations
- developer tools
- internal platforms
- AI-native UX
- structured memory
- retrieval pipelines
- vector search
- evaluation systems
- observability
AI workspace for structured memory, knowledge retrieval and personal workflows.
Highlights
- RAG memory
- knowledge graph
- vector search
- AI chat
- structured context
Code: Shadow-AI-Second-Brain
Turns Telegram work discussions into reviewed, structured Work Items.
Highlights
- Telegram capture bot
- LLM extraction → Work Items
- human review workflow
- FastAPI + async Python worker
- Postgres · Redis · Next.js
Code: telegram-task-bot
Production evaluation framework for LLM applications.
Highlights
- LLM-as-a-judge
- regression testing
- grounding
- safety evaluation
- human review
Code: ai-evaluation-tool
Operating system for managing AI agents, workflows, memory and project execution.
Highlights
- multi-agent orchestration
- agent lifecycle
- project memory
- execution tracking
- workflow routing
Prompt versioning, diffing, and rendering for LLM applications.
Highlights
- prompt versioning
- diffing and rendering
- structured prompt management
- monorepo tooling (Hono · Drizzle · Next.js)
Code: promptops-tool
- AI Product Engineering
- AI Agents & Workflows
- RAG & Memory Systems
- Developer Experience
- LLM Evaluation
- Automation Platforms
- AI Infrastructure
TypeScript · Python · React · Next.js · Node.js · PostgreSQL · Supabase · Azure · Vercel · OpenAI · Anthropic
6 years building and delivering B2B SaaS products.
Started as an implementation engineer and grew into technical leadership, coordinating engineers, QA, analysts, DevOps and business stakeholders while shipping production software for enterprise customers.
Today I focus on AI-native software, combining product thinking with engineering to build practical AI systems.
- AI-native products
- MCP
- Agent workflows
- Evaluation pipelines
- Long-term memory
- AI observability
I enjoy building developer tools, AI infrastructure, and experimental projects around agents, memory, and evaluation.