Experience
2024 - Present
Senior Machine Learning Engineer
Loka
Building AI systems on AWS infrastructure.
2023 - 2024
Co-founder
uxo.ai
Developed agents that understand and navigate the web.
2023
Machine Learning Engineer
Aimino
Implemented computer vision solutions robust against distribution shifts.
2022
Research Assistant
Explainable Machine Learning
Explored catastrophic forgetting in foundation models.
2020-2022
Graduate Student
University of Tübingen
Obtained a Master's degree in Machine Learning with GPA 1.0 (best: 1.0).
Open-source
I am a passionate proponent of open-source, and have contributed to several libraries:
Projects
Sometimes, I build stuff:
- Word Game Bench – evaluating language models on puzzles. OpenRouter sponsored the project, and Mark Chen (SVP of Research at OpenAI) expressed interest in the results.
- Answers to Chip Huyen's ML Interview Questions – a booklet answering interview questions covering Math, Computer Science, ML workflows and algorithms.
- Laser Hockey – my winning entry to a Reinforcement Learning tournament with 70 participants, organized by the Max Planck Institute for Intelligent Systems.
- Morty – An "agent" based on IBM Watson intents and an LSTM pre-trained on Reddit data I built all the way back in 2018!
Blog
Other times, I write stuff:
- Speeding up decoder inference with a Key-Value (KV) cache
- Boosting transformer efficiency with Grouped-Query Attention (GQA)
- Stabilizing training and improving model convergence with RMSNorm
- Activating neurons with Gated Linear Units (GLU) and Friends
- Encoding positional information with Rotary Position Embeddings (RoPE)
- Decoupling weight decay with AdamW
- Overcoming catastrophic forgetting with Elastic Weight Consolidation (EWC)
Biomarkers
With the goal of keeping myself accountable and maintaining good physical health, I'll continually open-source my biomarkers.