📍 Berlin, Germany
I build and operate production systems across infrastructure, reliability, data, and machine learning.
My work focuses on the engineering around intelligent systems — data pipelines, deployment, observability, evaluation, infrastructure, and production reliability.
LibreSRE YodaX Building SoloAI · Agentic AI · Cloud(AWS/Azure/GCP)· MLOps · AI Observability · SRE · Reliable AI Systems · Observability
📈 YodaX
Time-Series Forecasting & AI-Assisted Market Research
YodaX forecasts the next trading day's closing price for six US stocks using Google TimesFM 2.5 and evaluates saved predictions against actual market closes.
An experimental learning layer uses completed predictions to adjust future estimates without retraining the underlying model.
Future Market extends the project with a Gemini-powered commodity research assistant that analyzes recent news, follows unresolved questions, and builds linked evidence maps.
Time-Series Forecasting · Evaluation Pipelines · Prediction Tracking · Feedback Loops · TimesFM · Gemini
🤖 SoloAI
AI Infrastructure & Agent Control Plane
A control plane for connecting applications to permission-aware AI agents while keeping application infrastructure independent.
Focuses on multi-tenant architecture, model access, agent permissions, observability, evaluation, and infrastructure automation.
AI Infrastructure · Agentic Systems · Multi-Tenancy · Model Gateways · Observability · IaC · Azure
🚨 LibreSRE
AI-Assisted SRE Incident Investigation
Explores how AI can assist engineers during production incidents by correlating telemetry, surfacing hypotheses, and supporting structured troubleshooting while keeping humans in control.
SRE · Observability · Incident Response · Telemetry · Agentic AI · Root Cause Analysis
Explainable ML for Energy Systems
Anomaly detection for electricity consumption data using classical machine learning, deep learning, and explainability techniques.
PyTorch · CNNs · Autoencoders · Isolation Forest · LOF · SHAP · LIME
| Area | Technologies |
|---|---|
| Languages | Python · Go · SQL · Bash · Java |
| Cloud | AWS · Azure · GCP |
| Infrastructure | Kubernetes · Docker · Terraform · OpenTofu · Linux |
| Observability | Prometheus · Grafana · Datadog |
| Data | PostgreSQL · Databricks · Pandas · Polars · ETL/ELT |
| ML / AI | PyTorch · Scikit-learn · LLMs · RAG · Agentic AI |
| Backend | FastAPI · REST APIs |
| CI/CD | GitHub Actions · GitLab CI |
My background spans enterprise infrastructure, HPC, Linux, networking, cloud systems, and Site Reliability Engineering.
Today, I work across the intersection of:
Site Reliability Engineering · Platform Engineering · MLOps · ML Infrastructure · Data Engineering · Cloud Infrastructure
I'm particularly interested in the engineering required to make data and ML systems observable, scalable, explainable, and reliable in production.
Build intelligent systems that humans can understand, operators can trust, and infrastructure can reliably run.
Turning machine learning experiments into systems you can actually operate.