Building the machine a model actually wants
A model, once you stop thinking of it as math and start thinking of it as a machine, is a
Dave Lemphers
Latest
Writing our own inference engine in Rust on the AMD MI355X
On the small all-reduce that tensor-parallel inference runs for every token, our from-scratch Rust inference engine comes out about 1.3× faster than AMD's RCCL at its best, sitting on the physical floor of the MI355X.
What it's like being an AI Researcher on the Matilda team
We interview a lot of candidates who are making the move from academia into industry. It's a big
Learning to be Maincode
Why operationalizing discovery is the hardest and most important challenge ahead
Moving Beyond Static AI: The Need for Dynamic Personalisation and Where We Go Next
Why the one-size-fits-all approach to interaction modelling falls short and why we need to re-think the role of the human expert.
Why We Start Every Interview With “It’s 2 AM in the Lab”
We don’t hire for what you know today. We hire for who you are, and it starts with your most unforgettable 2 AM in the Lab moment.