Martin Noetzel
Ai Engineering Program Systems & Applied Research @Turing College
Signup · Get unlimited contacts
WORK HISTORY
Ai Engineering Program Systems & Applied Research @Turing College
Built a large-scale knowledge system (400+ books, PDF/EPUB) with hybrid retrieval (dense + sparse + RRF), cross-encoder re-ranking, and evidence-grounded answers with source citationsArchitected a privacy-first multi-agent system (LangGraph, MCP, Pydantic) with dual-domain isolation, deterministic execution graphs, and bounded autonomy (approval gates, tool allowlists)Implemented hybrid LLM routing across Anthropic (Claude), OpenAI, and local models (Ollama) optimizing cost, latency, and privacyDeveloped evaluation frameworks (golden datasets, RAGAS-style metrics) to measure retrieval precision, generation faithfulness, and decision qualityBuilt async-first full-stack system (FastAPI, PostgreSQL/pgvector, React/TypeScript) with OAuth2 auth, LLM tracing (Langfuse), and E2E test coverage (Playwright, 2,300+ tests)Tech:Python, FastAPI, PostgreSQL (pgvector), SQLAlchemy, LangGraph, Pydantic, Anthropic API (Claude), OpenAI API, Ollama, sentence-transformers, Langfuse, React, TypeScript, Docker, Playwright
EDUCATION
Hamburg University of Technology
Dipl. - Ing. oec., Industrial Engineering and Business Administration
SKILLS
ABOUT MARTIN NOETZEL
10+ years in consulting and transformation. I turn strategy into measurable execution — aligning operations, data, and performance management to improve productivity, scalability, and reliability in complex environments. I lead operational excellence and transformation initiatives, partnering with senior leadership to improve performance — for example by reducing cycle times, increasing productivity, and improving planning and delivery reliability. In parallel, I design and deploy AI systems for decision support — including large-scale knowledge retrieval (400+ sources) and controlled multi-agent architectures — to augment how organizations operate and make decisions. Most AI initiatives fail not from model capability, but from missing evaluation, integration, and operational reliability. I focus on that gap — ensuring solutions are embedded in real workflows and deliver measurable outcomes
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.