Senior AI & Generative AI Engineer / Researcher (PhD)
Architecting enterprise-grade Generative AI, scalable RAG pipelines, and agentic workflows. Bridging cutting-edge open-source research with resilient corporate backend infrastructure.
Selected Work
Open-source frameworks and applied research spanning vector search, foundation-model pretraining, and real-time AI systems.
Open-Source Vector Quantization Framework
High-performance Python library implementing advanced vector quantization. Dramatically reduces RAM and server footprint for large-scale semantic search and enterprise RAG without sacrificing retrieval accuracy.
Foundation Model Pretraining
Principal Investigator of the EU AI Factory project training a transparent, honesty-calibrated retrieval foundation model on European high-performance supercomputing infrastructure.
Hover to play demo Domain-Specific Evidence-Based RAG
Specialized RAG framework engineered for physiological and medical datasets. Implements hybrid semantic retrieval and deterministic guardrails for hallucination-free querying of clinical scientific literature.
Real-Time Conversational Voice AI — R&D Prototype
Full-stack conversational AI prototype exploring the latency boundaries of real-time voice agents. Evaluates speech-to-speech pipelines, WebSocket streaming architectures, and Time-to-First-Token (TTFT) optimization.
What I Build With
A full-stack toolkit for taking Generative AI from prototype to resilient production deployment.
Designing autonomous, tool-using agents and agentic workflows for enterprise automation.
Building retrieval systems that scale — accurate, auditable, and cost-efficient.
Resilient, cloud-native services that support high concurrent traffic.
A PhD-grade foundation in rigorous quantitative analysis of complex data.
Academic & Leadership
Two decades bridging computational neuroscience, high-dimensional data analysis, and applied AI engineering.
2021 — Present
Senior AI / Generative AI Engineer
Easy Software AG
EuroHPC Principal Investigator; enterprise RAG, agentic workflows, and scalable AI platforms.
2014
Habilitation & Senior Neuroscientist
University of Lübeck
2007
PhD — Magna cum laude
Max Planck Institute for Human Cognitive & Brain Sciences, Leipzig
2010 — 2013
Post-Doc Researcher
Helen Wills Neuroscience Institute, UC Berkeley, USA
EuroHPC Fast Lane
Principal Investigator — 27,000 NVIDIA A100 GPU-hours on CINECA Leonardo Booster (CANDOR).
DFG — German Research Foundation
Competitive research grant (€238,000).
Leopoldina — German Academy of Sciences
Post-doctoral research grant (€124,000).
Contact
Whether you are building enterprise RAG, exploring agentic systems, or investigating foundation-model research — I would be glad to talk.
contact@bahlmann.ai