Blog
February 18, 2026
Roanak BaviskarMarch 17, 2025
Grace BoyleFebruary 24, 2025Roanak Baviskar
What Neptune Got Right (And How to Keep It)
Neptune is shutting down on March 5, 2026 (10am PST). For teams that have been using it for years, the real loss isn't just the migration work. It's losing one of the few trackers that stayed fast when projects got big.
Roanak BaviskarPreflight Checks for GPU Fabric before Multi-node Training
Given the scale of modern LLMs, having a performant GPU network fabric is one of the most important factors for your training speed.
Grace BoyleGPU Utilization is a Misleading Metric
The most common metric ML teams use to understand GPU usage is GPU Utilization, but it can be misleading — you can hit 100% utilization while doing zero computations. Here's what to track instead.
Roanak Baviskar