Mischa Knabenhans
Senior Data Scientist
Zürich Metropolitan Area Switzerland
Summary
👤
Senior🎓
Top SchoolMischa Knabenhans is a Senior Data Scientist based in Zürich with eight years of experience applying advanced analytics, AI, and computational science to real-world problems. He holds a PhD in computational cosmology from the University of Zurich and built a highly efficient AI emulator (EuclidEmulator) used in astrophysics, demonstrating an uncommon mix of HPC, statistical UQ, and production-ready C++/Python tooling. In industry roles at UBS he automated risk workflows with deep learning and fine-tuned LLMs for classification, cut false positives in alerting systems by 50%, and turned unstructured email data into structured, mineable datasets. He combines hands-on engineering—ELT pipelines, webscrapers, SQL and Pandas—with clear stakeholder communication and project leadership in both public-private initiatives and sustainability-focused projects. Mischa pairs curiosity-driven research instincts with a pragmatic focus on delivering measurable business impact, and has recently augmented his toolkit with design thinking training from Harvard Business School Online. Notably, he bridges academic-grade model validation (<1% deviation in his emulator) with scalable production deployments in enterprise settings.
8 years of coding experience
Master of Science (MSc), Physik, 5.33, Master of Science (MSc), Physik, 5.33 at Eidgenössische Technische Hochschule ZürichMatura, 5.56, Matura, 5.56 at Kantonsschule Freudenberg (KFR)Certificate of Completion, Design Thinking and Innovation, passed, Certificate of Completion, Design Thinking and Innovation, passed at Harvard Business School OnlineDoktor (Ph.D.), Computational Science (focus: computational cosmology), Doktor (Ph.D.), Computational Science (focus: computational cosmology) at Universität Zürich | University of Zurich
German, English, Spanish, French
Github Skills (14)
zx-spectrum10
estimate10
spectrum10
google9
signal-processing9
python9
mercury9
hpc-applications9
renewable-energy9
predictor9
certificate8
automation8
science7
machine-learning5
Programming languages (2)
Github contributions (5)
miknab/EuclidEmulator2
Sep 2020 - Mar 2021
EuclidEmulator2 is the updated version of EuclidEmulator, a fast and accurate AI-based predictor for the non-linear correction of the matter power spectrum.
Contributions:2 releases, 28 commits, 8 PRs in 6 months
pythonmachine-learninglinearnon-linearpower-spectrum
miknab/PlanetsEmulator
Nov 2018 - Jan 2020
Contributions:128 pushes, 1 branch in 1 year 1 month
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.