Biography
I’m a senior machine learning engineer with a research background. My work concentrates on three areas:
- Production ML systems: training + evaluation pipelines, feature/data quality, reproducibility.
- Performance & cost: profiling, workload consolidation, pragmatic refactoring of critical paths.
- Reliability & observability: clear metrics, failure analysis, reducing variance in model behaviour and deployments.
Previously, I did research in machine learning and feature relevance with a focus on reproducibility and efficient experimentation (high‑performance computing roots). I still care about lean experiment design, versioned data, and traceable results.
Selected themes I keep returning to:
- Simplifying overgrown internal platforms instead of adding another layer.
- Automating the boring parts of evaluation so human review time is spent on edge cases.
- Treating model + data + infra changes as one unit when reasoning about regression risk.
Outside work: strength training, mountain trails, strategy / systems games, and the occasional attempt to cook something plant‑based that isn’t another lentil variant.
- Advanced Machine Learning Techniques
- Cutting-edge Software Engineering Practices
- Machine Learning Operations (MLOps)
- Innovative Cloud Application Development
-
Dr. rer. nat. Machine Learning, 2021
Bielefeld University
-
B.Sc & M.Sc. in Bioinformatics and Genome Research, 2014 & 2016
Bielefeld University
Recent Posts
Experience
- Document Intelligence: Enhanced product features for improved document analysis, focusing on customized solutions for client-specific needs. Developed advanced techniques for efficient and accurate extraction of key features from diverse document types.
- Backend Optimization: Led the optimization of the backend platform to manage large-scale data efficiently. Focused on key metrics like cost, quality, and speed, achieving significant improvements in handling customer workloads. Implemented enhancements to the processing flow, drastically reducing error rates in the overall pipeline by 100x. Worked on the overall observability platform, establishing key technical metrics that guided further improvements.
- Technical Leadership: Provided guidance and expertise to the team, fostering a culture of agility and continuous improvement. Conducted research and implemented new software tools to accelerate development and improve system maintainability.
- ML Engineering: Designed a scalable AI model backend service (DB, API) using an event-driven serverless architecture. Established a Kubernetes-based model deployment, reducing costs by 80%. Developed and extended open-source Python SDKs for interacting with the platform.
- Cloud Engineering: Led the migration of a heterogeneous legacy AWS stack to Kubernetes using Terraform (IaC), resulting in a 60% cost reduction, enhanced fault tolerance, and streamlined maintenance. Established compliance rules within the organization (SOC2).
- DevOps: Maintained and expanded CI/CD pipelines (Jenkins, ArgoCD) for backend and application deployments, and introduced coding standards to increase agility.
- AI Research: Investigated and implemented AI solutions to address customer challenges, utilizing state-of-the-art large language models (LLM) for text matching and information extraction applied to millions of records.
Responsibilities include:
- Research ( Modelling & Analysis )
- Development and Deployment
- Scientific writing and presentation
Responsibilities include:
- Research ( Feature Representation, Non-linear Models)
- Scientific writing and presentation
Responsibilities include:
- Research & Development
- Scientific writing and presentation
- Teaching
Project Goal: Android App for medical professionals handling and washing endoscopes in a clinical setting.
Developed in competitive setting with the endoscope manufacturer (Miele) as client. Focus on agile development with changing requirements.
Responsibilities include:
- Software architecture
- Task delegation
- DevOps
- Development
