Andreas Madsen is a Machine Learning Engineer and founding research scientist with 15 years of experience bridging industry-grade engineering and academic research in NLP interpretability. He holds a PhD from Mila and focuses on making LLM chain-of-thought and self-explanations verifiable and actionable, grounding methods in practical deployment needs. A long-time open-source maintainer and official Node.js collaborator, his early contributions to core modules like child_process and cluster helped shape production JavaScript runtime behavior and his packages saw tens of millions of downloads. He has led teams and shipped products from Clinic.js profiling tools to recruiter search engines, and his ICLR spotlight award and Microsoft internship reflect research that translates into product impact. Based in Chiyoda, Tokyo, he combines deep mathematical training with a pragmatic engineering lens—work you can explore at his portfolio.
14 years of coding experience
8 years of employment as a software developer
Doctor of Philosophy - PhD, Computer Engineering, NLP Interpretability, Doctor of Philosophy - PhD, Computer Engineering, NLP Interpretability at Mila - Quebec Artificial Intelligence Institute
Doctor of Philosophy - PhD, Computer Engineering, 4, Doctor of Philosophy - PhD, Computer Engineering, 4 at Polytechnique Montréal
Bachelor's degree, Mathematics and Technology, Grade: 98%, Bachelor's degree, Mathematics and Technology, Grade: 98% at Technical University of Denmark
Gymnasium HTX, Physic and Mathematics, Gymnasium HTX, Physic and Mathematics at Ørestad Gymnasium
Contributions:7 reviews, 83 PRs, 26 pushes in 8 years 2 months
Contributions summary:Andreas primarily contributed to the `nodejs/node` repository, focusing on improvements and additions to the child process and cluster modules. Their work included fixing bugs, adding new features like the "silent" option to `child_process.fork`, and adding an `env` argument to `cluster.fork`. The user's contributions also extended to improving test cases and refactoring internal code within the cluster module.
Contributions summary:Andreas focused on improving the `htmlparser2` library's core parsing logic. Their commits primarily addressed issues in the tokenizer, specifically correcting how special tags, comments, and CDATA sections were handled. They optimized the parser by moving if statements into methods, allowing for potential performance improvements. Additionally, the user made changes to the parser to handle implicit tag openings correctly.
xmljavascripthtml-parserxml-parserparser
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.
Request Free Trial
Andreas Madsen - Machine Learning Engineer at guidelabs