AISSAI-JAXHACK : JAX for HPC/AI hybridization hackathon

Hackathon – JAX for HPC/AI Hybridization
September 7–9, 2026 • IDRIS, Paris-Saclay

The JAX for HPC/AI Hybridization hackathon is a three-day hands-on event dedicated to evaluating JAX for large-scale scientific computing and the convergence of High-Performance Computing (HPC) and Artificial Intelligence (AI).

Participants will explore JAX’s strengths and limitations through hands-on exercises, scientific computing applications, and scalability studies conducted on the Jean Zay supercomputer.

Two Participation Tracks :

Track 1 – JAX for Scientific Computing (Beginner Level)

This track is designed for researchers, engineers, and developers with experience in scientific computing, HPC, or numerical methods who are new to JAX.

Participants will be able to:

Learn JAX fundamentals through intensive “crash courses” each morning;
Apply their new skills to scientific computing use cases;
Build reproducible benchmarks to compare JAX against standard scientific frameworks.

Track 2 – Pushing the Limits of JAX (Advanced Level)

This track is designed for experienced JAX users wishing to explore its performance and scalability on HPC systems.

Participants will be able to:

Work on real-world JAX applications throughout the event;
Investigate multi-GPU and multi-node scaling on Jean Zay;
Identify bottlenecks, missing features, and optimization opportunities;
Produce advanced performance and scalability studies.