Fabio Rocha
Portfolio Manager at qSpark Capital Management
Porto, Porto, Portugal
Summary
🤩
Rockstar🎓
Top SchoolFabio Rocha is a portfolio manager and quantitative researcher based in Porto with a PhD in Physics from Princeton and over a decade of experience building systematic trading systems and HFT strategies. He has progressed from hands-on quant and HFT roles to opening a European office and leading portfolio management at qSpark Capital, blending research-grade modeling with production trading. Fabio also spent time as a Staff Software Engineer contributing to PyTorch—adding tensor operations, reference implementations, tests, and performance optimizations with Triton—demonstrating rare fluency across ML engineering and low-latency finance. He combines deep theoretical training with pragmatic software craftsmanship, routinely moving ideas from prototype code into live trading or open-source commits. Colleagues value his ability to straddle research, engineering, and trading operations while shipping robust, performance-sensitive systems.
3 years of coding experience
10 years of employment as a software developer
Licenciatura, Physics, Licenciatura, Physics at Universidade do PortoPhD, Physics, PhD, Physics at Princeton University
Github Skills (13)
tensorrt10
pytorch10
machine-learning10
triton10
tensor10
deep-learning10
tensorflow10
operation10
python10
numpy9
testing9
autograd8
gpu8
Programming languages (3)
Github contributions (5)
pytorch/pytorch
Jul 2022 - Jan 2023
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:71 reviews, 234 commits, 62 PRs in 6 months
Contributions summary:Fabio primarily contributed to the PyTorch library, focusing on adding and improving functionality related to tensor operations and deep learning components. Their work included implementing new operations like `unflatten` and `unbind`, enhancing existing functions like `glu`, and adding OpInfo for these and other functions to enable testing. The user also added or updated reference implementations for several functions, contributing to the robustness and completeness of the library. Furthermore, the user demonstrated expertise in performance optimization by implementing optimizations using triton.
pythongpu-accelerationdeep-learninggpunumpy
fdrocha/triton
Oct 2022 - Feb 2023
Development repository for the Triton language and compiler
Contributions:13 pushes, 2 branches in 4 months
compilerpreprocessorssaprogramming-languagetriton
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.