Viktor Gal
Lead AI ML Engineer at Pruplebird
Zurich, Zurich, Switzerland
Summary
🤩
Rockstar🎓
Top SchoolViktor Gal is a Lead AI/ML Engineer based in Zurich with 17 years of experience blending academic rigor and production-grade engineering across research labs, startups, and enterprise teams. He holds a PhD in Applied Mathematics and Computer Science and has driven ML systems from postdoctoral research at ETH Zurich to senior roles at the Swiss Data Science Center and DataRobot. Viktor combines quantitative finance work—developing investment strategies as a Quantitative Analyst—with hands-on ML engineering, including optimizing kernels in the well-known Shogun toolbox. He is comfortable moving between low-level performance improvements and high-level algorithm design, and often surfaces small but impactful optimizations (e.g., algorithmic tweaks that replace divisions with multiplications) to boost efficiency. Known for bridging research and product delivery, he thrives on turning complex models into reliable, performant systems.
17 years of coding experience
14 years of employment as a software developer
Australian National UniversityMaster’s Degree, Computer Science, Master’s Degree, Computer Science at Budapest University of Technology and EconomicsDoctor of Philosophy (Ph.D.), Applied Mathematics and Computer Science, Doctor of Philosophy (Ph.D.), Applied Mathematics and Computer Science at Ghent UniversityExchange, Computer Science, Exchange, Computer Science at Teknillinen korkeakoulu-Tekniska högskolan
English, Serbian, German, Hungarian
Github Skills (8)
kernel10
machine-learning10
c-language10
cprogramming-language10
kernel-mode10
data-science9
mathematics8
math8
Programming languages (20)
Github contributions (5)
shogun-toolbox/shogun
Feb 2012 - Dec 2020
Shōgun
Role in this project:
Data Scientist & ML Engineer
Data Scientist & ML EngineerContributions:7 releases, 1 review, 1536 commits in 8 years 11 months
Contributions summary:Viktor's contributions focused on implementing and optimizing a Jensen-Shannon kernel within the Shogun toolbox, adding a CDotKernel-based variant. They further refined the kernel's performance by changing to CMath::log2 and optimizing the computation using a multiplication instead of two divisions. Their work involved extending the modular interface by including the Jensen-Shannon kernel definition and modifying various header files.
cmakedata-sciencegunc-plus-plusmachine-learning
vigsterkr/netket
Jun 2020 - Feb 2025
Machine learning algorithms for many-body quantum systems
Contributions:15 pushes, 7 branches in 4 years 8 months
quantum-computingmachine-learning-algorithmsquantum-many-bodybodylearning-algorithms
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.