Production
Working on data products and process analytics
At dmTECH I work on data products, process analytics, and the pipelines, internal packages, and deployments around them across Python, PySpark, Docker, Terraform, and Google Cloud.
Machine Learning / Applied AI / Data Science
Joscha Lasse Bisping
Master's student at TU Berlin working on my thesis, graph neural networks and LLM-derived embeddings for ransomware detection. I build data products at dmTECH, and I'm exploring an early-stage idea around digital twins for critical infrastructure.
- Current role
- Working Student Data Science
- Based in
- Berlin, Germany
- Education
- Master of Science,
Technische Universität Berlin
Trajectory
Work, research, and teaching.
Research
Combining graph ML with LLMs
My current research explores how LLM-derived labels, embeddings, and related signals can enhance graph structures and graph learning workflows.
Teaching
Teaching computer science
I taught algorithms and programming at TU Berlin and learned how to explain technical concepts clearly and concretely.
Selected work
Projects that sit at the intersection of ML, systems, and applied problem solving.
Project
Tsunami Agent
LangChain-based LLM agent to autonomously generate, compile, and debug Java plugins for Google's Tsunami Security Scanner. Designed a two-stage workflow combining code generation with compiler feedback for self-correction, enabling fully automated plugin development for OWASP Juice Shop vulnerabilities. In evaluation, the agent produced 16 working vulnerability detection plugins, all without manual code fixes.
- Python
- LangChain
- AI/ML
- Cybersecurity
- Google Tsunami
Project
Satellite Image Segmentation Pipeline
Geospatial Computer Vision pipeline for automated building detection from multispectral satellite imagery. Integrated Sentinel-2 and OpenStreetMap data, generated segmentation masks, prepared stratified training/validation datasets, and trained PyTorch-based CNN and U-Net models with hyperparameter tuning and augmentation.
Project
Train Stop Detection from Phone Sensors
Project to detect station stops of Berlin commuter trains using only phone sensor data that apps can access without special permissions (magnetometer, accelerometer, gravity, and gyroscope streams). The data is resampled and transformed into time-based features that are less sensitive to how the phone is oriented and used to train an ML model that predicts stop intervals for each ride. Uses HistGradientBoosting with 328 engineered features including lag and diff features. Achieved 0.722 Jaccard score and top leaderboard position.
Technical focus
Technical skills and tools.
I work most comfortably where machine learning, data engineering, and product-minded implementation meet. For a fuller picture, the CV page covers coursework, teaching, volunteering, and the broader toolset I worked with.
Programming Languages
- Python
- JavaScript
- Java
- C
- SQL
Machine Learning & Data Science
- PyTorch
- PyTorch Lightning
- PyTorch Geometric
- Graph Neural Networks (GNNs)
- Large Language Models (LLMs)
Data Visualization
- Seaborn
- Matplotlib
- Plotly
Open to roles in Machine Learning / AI Research, applied ML, and technically demanding product work.
If the role needs someone comfortable moving between research context, implementation, and explanation, that is the space I like working in. I'm also open to early-stage collaborations.