Next 100x in AI: Inference, Networking, & Self-Optimizing Models — Philip Kiely & Ali Taha, Baseten

Describes a concrete loop where a model profiles a forward pass, finds bottleneck kernels and writes replacements, now used in a production inference engine.

• inference — Running a trained model to get answers — the phase where AI is actually used, as opposed to trained.

• videoLong Live Context Engineering - with Jeff Huber of Chroma
• videoThe AI Frontier: from Gemini 3 Deep Think distilling to Flash — Jeff Dean
• videoGreg Brockman on OpenAI's Road to AGI
• video🔬 Why Transformers Hit a Wall the Moment Physics Shows Up — Anima Anandkumar, Caltech

• videoThe Inference Frontier: from 100 to 10,000 tokens per second — Sean Lie, Cerebras CTOLatent Space
• videoWhat's New in Inference Engineering — Philip Kiely, BasetenAI Engineer
• podcastAI researchers debate how close we are to recursive self-improvementDwarkesh Patel

Checking sign-in…

Loading comments…