Jakob Kruse
Supervision Analyst at appliedAI Initiative GmbH
Frankfurt, Hesse, Germany
Summary
👤
Senior🎓
Top SchoolJakob Kruse is an AI and machine learning specialist and software engineer with eight years’ experience building production-focused ML systems and tooling across academia and industry. He combines deep technical skills in Python and C++ with hands-on experience deploying LLM and generative AI products as a product owner and project coordinator at the European Central Bank and appliedAI. His background spans TUM and MIT research, contributions to open-source tooling for invertible architectures (FrEIA) including a Gaussian Mixture Model module, and practical optimization work at Infineon that yielded substantial kernel speedups for microcontrollers. Jakob is comfortable moving between research, engineering, and product roles—scoping LLM/RAG solutions, leading small teams, and shipping containerized components. He is finishing a PhD in machine learning while mentoring and growing university AI initiatives, reflecting both academic rigor and a knack for building communities. An often-overlooked strength is his ability to translate low-level numerical optimization into scalable ML products for regulated environments.
8 years of coding experience
2 years of employment as a software developer
Visiting Student Researcher, Visiting Student Researcher at Massachusetts Institute of TechnologyMaster of Science - MS Electrical Engineering and Information Technology, Master of Science - MS Electrical Engineering and Information Technology at Technical University of MunichA-levels / Abitur, A-levels / Abitur at Gymnasium Oberhaching
English, German, French, Spanish
Github Skills (5)
pytorch10
machine-learning10
python10
web-framework9
testing8
Programming languages (1)
Github contributions (5)
vislearn/FrEIA
Mar 2020 - Apr 2021
Framework for Easily Invertible Architectures
Role in this project:
ML Engineer
ML EngineerContributions:9 reviews, 29 commits, 3 PRs in 1 year 1 month
Contributions summary:Jakob primarily contributed to the development of a Gaussian Mixture Model (GMM) module within the FrEIA framework. This module includes functionalities for defining invertible GMM architectures, handling component weights, means, covariance parameterization, and mixture component selection. The user implemented forward and backward passes for the GMM, along with Jacobian calculations, enabling its use within the broader invertible neural network framework. Additionally, the user created and fixed test scripts.
architecturesopenflowframeworkdata-mesharchitecture
vislearn/HINT
Mar 2020 - Feb 2021
Code for the research paper "HINT: Hierarchical Invertible Neural Transport for Density Estimation and Bayesian Inference".
Contributions:11 commits, 9 pushes, 1 branch in 11 months
autoencoderbayesian-inferenceinferencehierarchicaldensity-estimation
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.