Stories by robertnishihara

Major upgrades to Ray Serve: 88% lower latency and 11.1x higher throughput

SkyRL brings Tinker to your GPUs (2025)

vLLM large scale serving: DeepSeek 2.2k tok/s/h200 with wide-ep

Massively Parallel Agentic Simulations with Ray

Deploy DeepSeek‑R1 with VLLM and Ray Serve on Kubernetes

An Open Source Stack for AI Compute: Kubernetes and Ray and PyTorch and VLLM

Native LLM APIs in Ray Data and Ray Serve

Joins and Hash-Shuffle in Ray Data

AsyncFlow: An Asynchronous Streaming RL Framework for LLM Post-Training

Open Source RL Libraries for LLMs

Large-Scale Deployment of Ray in Tencent's Weixin AI Infrastructure

Uv and Ray: Pain-Free Python Dependencies in Clusters

Roll: Reinforcement Learning Optimization for Large-Scale Learning

An Open Source Stack for AI Compute: Kubernetes and Ray and PyTorch and VLLM

Uv and Ray: Pain-Free Python Dependencies in Clusters

Ray Batch Inference at Pinterest (Part 3)

Direct Preference Optimization with Synthetic Data on Anyscale

Building an LLM Router for High-Quality and Cost-Effective Responses

Ray Infrastructure at Pinterest

1 2