BusyJay

Jay

I shave all those, and those only, who do not shave themselves.

Running a 95 GiB MoE Model on a 32 GiB GPU: From 10 to ~100 Tokens/s

Oct 26, 2026 Running a 95 GiB MoE Model on a 32 GiB GPU: From 10 to ~100 Tokens/s

Extending Raft algorithm: Witnesses and Arbiters, Part 2

Sep 19, 2026 Extending Raft algorithm: Witnesses and Arbiters, Part 2

Extending Raft algorithm: Learner, Part 1

Aug 2, 2024 Extending Raft algorithm: Learner, Part 1

How to implement reliable and efficient leader lease

Apr 11, 2024 How to implement reliable and efficient leader lease