Content
SB AI Labs
by Simon Budziak — CTO · Founder · AI Engineer · LangChain Ambassador
I help teams ship production LLM agents that work under real users, not just in demos.
AI Agents · LLMs · LangChain · LangGraph · Claude Code · Evals & Observability · LLMOps
What I Offer
How I can help you
Pick the engagement that fits your situation — or reach out and we'll figure it out together.
AI Agent Systems
End-to-end production LLM agent pipelines for enterprise teams. I design, build, and harden systems that don't break under real users.
- →RAG pipelines with retrieval that actually works
- →Multi-agent orchestration with LangGraph
- →LangSmith evals & observability from day one
- →Production deployment & on-call handoff
Freelance AI Engineering
MVP to production. I join your team as a senior AI engineer — POC validation, architecture review, or full-stack LLM feature delivery.
- →POC → production transition plan
- →LLM architecture & stack selection
- →Code review & evals setup
- →Weekly async or sync collaboration
AI Assisted Coding Workshop
Hands-on 6-hour workshop for tech teams to ship faster with AI Assisted Coding, without compromising quality or security.
- →6-hour remote or on-site format tailored to your team
- →Hands-on tooling workflow with Claude Code, Cursor, and Codex
- →Security, governance, and best practices for enterprise adoption
- →Recording, practical materials, and post-workshop support
Speaking & Consulting
Conference talks, meetup keynotes, and strategic AI consulting for leadership teams navigating LLM adoption.
- →Production LLM systems talks
- →LangChain community events
- →AI strategy for CTOs & VPs
- →Team capability assessments

About
Building AI that ships
I'm a Senior AI Engineer and Co-Founder of Lubu Labs, based in Wroclaw, Poland · San Francisco, USA. Poland's first LangChain Ambassador, I specialize in production LLM agent systems — RAG pipelines, tool use, LangGraph orchestration, and LangSmith observability.
My work spans from NATO defense hackathons to YC-stage startups. I build systems that work reliably under real users, with evals and observability baked in from day one.
Stack
Experience
Building production AI agent systems for enterprise clients
Leading AI-powered game porting technology development
EasyGamePort (YC finalist), SynverAI, Doers.ai — shipped 3 AI products
NVIDIA Jetson + DeepStream + TensorRT → -40% inference time
Built HPCBursar and GSI-SSH Gem for supercomputing infrastructure
Companies I have worked with
Work
Projects
AI SDR Chatbot
Customer-facing sales agent with 100% production reliability. Built with LangGraph for stateful conversation management and tool use.
Healthcare Patient Agent
Patient education agent achieving 40% fewer false positives and 80% routing accuracy. Integrated with clinical workflows.
EasyGamePort
AI agent app to port games to consoles automatically. Reached Y Combinator final interview round.
LangChain Agent Skills
80+ ⭐ open-source collection of 9 production-ready skills for LangChain, LangGraph, and LangSmith AI coding assistants.
LangGraph Lightning Optimization
Prompt auto-optimization using APO algorithm with LangSmith tracing integration for systematic agent improvement.
VibeRadle
Daily puzzle game about global trade and world economies. Two modes: VibeTrace and VibeProbe. Built with vibe coding.
Contact
Let's Build Something
Whether it's a production agent system, a workshop for your team, or just a chat about LLMs — I'm always happy to connect.


