Philip Kelly

Person

Mentions

Episodes

Claims

Philip Kelly, mentioned 11 times across podcast episodes and expert conversations analyzed by Sonic.

What Philip Kelly has said

Philip Kelly's book, "Inference Engineering," posits that AI inference is a full-stack problem combining CUDA, on-GPU optimization, and distributed systems.

neutral·Philip Kelly

The GPT OSS 120B model is considered too large for many use cases, making smaller models like the Gemma family more suitable for fine-tuning.

bearish·Philip Kelly

Baseten was a day-zero support partner for the launch of Google's Gemma 4 model.

bullish·Philip Kelly

Philip Kelly's book, "Inference Engineering," posits that AI inference is a full-stack problem combining CUDA, on-GPU optimization, and distributed systems.

Expert perspectivePhilip KellyApr 28

The GPT OSS 120B model is considered too large for many use cases, making smaller models like the Gemma family more suitable for fine-tuning.

Expert perspectivePhilip KellyApr 28

Baseten was a day-zero support partner for the launch of Google's Gemma 4 model.

Expert perspectivePhilip KellyApr 28

Baseten operates a multi-region deployment that pools GPUs from America and Europe into a unified compute pool to minimize user latency.

Expert perspectivePhilip KellyApr 28

Baseten was one of the first companies to use NVIDIA Dynamo for inference in a production environment.

Expert perspectivePhilip KellyApr 28

The Gemma family of models from Google supports native image inputs, which is beneficial for enterprise use cases like KYC and document extraction.

Expert perspectivePhilip KellyApr 28

The low-latency networking of Google Kubernetes Engine (GKE) saves Baseten a couple dozen milliseconds per turn between models in multi-model agentic systems.

Expert perspectivePhilip KellyApr 28

Baseten utilizes Google Kubernetes Engine (GKE) to build its systems on top of Google Cloud GPUs.

Expert perspectivePhilip KellyApr 28

The Gemma family of models provides a wide range of sizes, from 2 billion to 30 billion parameters.

Expert perspectivePhilip KellyApr 28

Baseten offers NVIDIA's Nemotron models, such as Nemo Tron Super, through a pay-per-token API.

Expert perspectivePhilip KellyApr 28

Migrating an inference system from NVIDIA's Hopper architecture to the Blackwell architecture requires significant software and kernel layer work.

Expert perspectivePhilip KellyApr 28

Create a free account to see Philip Kelly's full intelligence report - every claim, the relationship network, and AI Q&A across all sources. No card needed.

Get started free

Back to Entities Entity Detail