Frimousse

Léopold Cambier (leo @ this website)
Senior Software Engineer, NVIDIA
Former Ph.D. Student, ICME, Stanford University


ORCID iD iconorcid.org/0000-0002-0147-086X

spaND’s public repo on Github

TaskTorrent’s repo on Github

Current Work

I am currently a Software Engineer at NVIDIA in the CUDA Math Libraries team.

Mini-Bio

I did my Bachelor & Master (2010-2015) in Engineering at Université catholique de Louvain (Belgium). During that time, I spent 3 months at MIT working with Prof. Laurent Demanet on seismic imaging and full wave inversion. I did my master thesis with Prof. P.-A. Absil, working on Low-Rank Matrix Compression.

In 2015 I moved to Stanford to start my Ph.D and started working with Prof. Eric Darve (Mechanical Engineering) on numerical linear algebra.
We worked on bridging the gap between direct methods (accurate, reliable but expensive) and preconditioners (cheap but complex to work with or very domain specific). We developed highly efficient incomplete factorization methods for large sparse linear systems, typically arising from PDE’s applications. And we combined this with parallel runtime systems. My thesis is entitled Fast and scalable hierarchical linear solvers. I graduated in January 2021. During my PhD, I did two interships (summers 2016 and 2017) at NVIDIA, working on cuDNN and cuBLAS. More recently I interned at Intel (summer 2019), working on reduce precision arithmetic for neural networks trainings.

I currently work as a software engineer at NVIDIA in the CUDA Math Libraries team. At NVIDIA I worked on FFT and GEMM libraries and Python CUDA compilers.

An up-to-date (as of September 2024) resume can be found here