- Published on
Evolving GPU Kernels
- Authors
- Name
- Martin Andrews
- @mdda123
Presentation Link
The sixty-first MeetUp of the Machine Learning Singapore Group, was titled : "LLMs for Software and Wetware" - and had a bit of a Nature twist : The evolution of programs; and using programs to help design medicines.
My Presentation
My talk was titled "Evolving GPU Kernels", and had the following outline:
- GPU Kernels
- What's involved?
- AMD Challenge *(Evolutionary Algorithms
- Some History / Ideas
- AlphaEvolve
- What's new?
- Wrap-up & QR-code (the latter to reduce audience distractions)
I had already been working with Evolution methods to create GPU kernels, as part of my involvement with the AMD Developer Challenge (2025). So this talk explained some of my choices - going back to the early days of GAs/GPs during my PhD in the UK.
Coincidentally (or perhaps because there's a lot of it in the air at the moment), the AlphaEvolve tech report from DeepMind was also released just prior to the MeetUp. That gave me a nice segue into describing what they were doing!
The slides for my talk, which contain links to all of the reference materials and sources, are here :

If there are any questions about the presentation please ask below, or contact me using the details given on the slides themselves.

Other Presentations
In his talk "A Deep Learning Approach for Nanomedicine Design", Alvin Chan explained his work at NTU with training Transformer-style models to classify (and hence select suitable experimental candidates) for Nanomedicine design. This was a very interesting talk, which our audience had been asking for. Hopefully we'll be able to host more AI-for-Science subjects in the future.
Acknowledgements
Many thanks to the Google team, who allowed us to use Google's Developer Space.