# Team Manifold - includes NVIDIA employee

- **Event:** [PyTorch Helion Hackathon](https://cerebralvalley.ai/e/helion-hackathon)
- **When:** Sat, Mar 14 at 9:00 AM – 10:00 PM (PDT)
- **Where:** Shack15, San Francisco, CA
- **Team:** [Josh Rauvola](https://cerebralvalley.ai/u/jrauvola), [Andrew Briand](https://cerebralvalley.ai/u/andrewbriand)
- **GitHub:** https://github.com/andrewbriand/helion-hackathon/tree/main/problems/helion/causal_conv1d_py
- **Gallery:** https://cerebralvalley.ai/e/helion-hackathon/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/helion-hackathon/hackathon/gallery/8

We removed all the extra compute in all the kernels. We adjusted tile size and autotuned the kernels and used hl.dot instead of separate outer products for the gated_deltanet_recompute_w_u_py.

---

Markdown version of https://cerebralvalley.ai/e/helion-hackathon/hackathon/gallery/8. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt