I build software. Currently Founder & CTO of Mantle. Side work lately is mostly LLM inference and agent infrastructure — what most of the recent posts here are about. Research background in computational physics and GPU-accelerated scientific computing — an early GPGPU practitioner from a decade before GPUs became the substrate of modern AI.
Previously: Senior Director of Engineering at Clio, co-founder of Evey Events (acquired 2022), and Director of Engineering at Shopify. M.Sc. Computer Science (computational), Memorial University.
Find me on GitHub, X, or email.
Recent posts
- Sep 20 A fluid simulation where Jev decides every collision
- May 30 Searching for tiny-cache LLMs
- May 17 Compressing the KV cache without retraining, and what mostly doesn't work
- May 15 Trying to push sparse attention further, and what didn't work
- May 13 90% of attention is slack in pretrained transformers