Martin Ulises Millán · M.Sc. Computer Science · Cuernavaca, México 🇲🇽
I build production AI systems end-to-end — from research papers to deployed products. My path: Mechatronics → AI pivot because code scales where hardware doesn't. M.Sc. in Computer Science (2025), guest researcher in Spain, currently shipping AI agents, RAG pipelines, and fullstack apps for enterprise clients.
Currently open to senior AI/ML or fullstack roles at global companies. Interested in: AI agent architectures · cloud-native systems · computer vision · RAG at scale
Real production projects shipped at Beecker (private repos, but the impact speaks for itself):
|
🤖 Slack Meeting Bot ·
|
💬 Enterprise RAG Chatbot Information-grounded chatbot for company knowledge, integrated with Linear MCP, deployed on Azure with full CI/CD.
|
|
📄 Document Validation API Playwright-based API that validates Mexican government IDs (SAT/INE) in real-time, returning status + reasons.
|
🌐 Fullstack WebApps End-to-end webapps: Figma MCP → Next.js + shadcn/ui frontend, FastAPI/Flask backend, Azure-deployed.
|
| Where | What | Result |
|---|---|---|
| 🎓 CENIDET (M.Sc., 2025) | GAN detection — real vs. synthetic face classification | >98% accuracy across 7+ benchmarks |
| 🔬 Universidade da Coruña 🇪🇸 (Guest Researcher, 2025) | Glaucoma screening from retinal images | 84% → 90% accuracy improvement |
| 👥 Beecker (2023–present) | Tech advisor on architecture, tooling, and team decisions | Sole IC owning backend, frontend, infra, and AI |
Languages · Python TypeScript JavaScript SQL C++
AI/ML · PyTorch OpenCV scikit-learn LangGraph RAG MCP CNN
Frontend · Next.js React shadcn/ui Tailwind
Backend · FastAPI Flask Node.js uvicorn
Data · PostgreSQL (pgvector) MongoDB Chroma Redis
Cloud / DevOps · Azure Docker CI/CD Linux Git
Automation · Playwright Selenium pywinauto pyautogui
- 🎯 Production AI agents — going past chatbots into real orchestration with context, memory, and tool use
- ☁️ Cloud depth — leveling up from Azure into AWS + Kubernetes for global-scale systems
- 📚 Open source — starting to contribute intentionally, not to look good on paper
- 🤝 Network — connecting with engineers at the companies I want to learn from
