ray

@GEOTAB Request access for the Optimization Services Program here: anyscale.com/asset/anyscale…

@GEOTAB, a leader in fleet management and safety analytics for long-haul drivers, debug complex video pipelines with multiple AI models. With Agent Skills, they move faster without manual log-diving or deep distributed systems expertise, enabling more teams to efficiently use AI to process and train models on their large volumes of video data. Blog: na2.hubs.ly/H05fHk30

Production VLA pipelines scale three workloads across three hardware profiles, plus thousands of stochastic sim rollouts. Join @anyscalecompute + @wandb May 12 to orchestrate it end-to-end with Ray. Register: na2.hubs.ly/H05dzc10

We're hosting a live webinar on architecting multimodal data pipelines that scale. You'll leave with a clear view of how Ray and Anyscale work together to keep your GPUs busy. If you are a computer vision engineer, researcher, or ML engineer working on physical AI or multimodal systems, this session is for you. Register: na2.hubs.ly/H051Z7w0

Want early access to the Optimization Services Program? Request access here: anyscale.com/asset/anyscale…

Setting up Anyscale on Kubernetes means aligning Helm, Terraform, IAM, networking, and GPU config together, each depending on the one before. Infra Skills guides your coding agent through the process and generates configs for the environment you actually run. Blog: na2.hubs.ly/H059t0S0

The Optimization Services Program is also available. Request access here: anyscale.com/asset/anyscale…

One prompt can now do more than generate code. With Anyscale Agent Skills, one skill can build the Ray batch inference pipeline and another can launch the job, generate the workspace config, and stream logs through completion. From prompt to running workload, in one flow. Blog: na2.hubs.ly/H0552NP0

In this session you will learn: - Why multimodal data pipelines break traditional batch processing engines and architectures - How to avoid I/O between CPU → GPU steps that lead to +50% GPU idle time - How to build a streaming, CPU+GPU pipeline with Ray Data Register: na2.hubs.ly/H051YYt0

Want access to Optimization Services Program? Request it here: anyscale.com/asset/anyscale…

When a Ray job fails in production, generic agents cannot inspect live cluster state or trace the failure to the real cause. Platform Skills read logs and metrics, diagnose the issue, patch the fix, redeploy, and verify the result in one conversation. Blog: na2.hubs.ly/H050GWY0

Want early access to Optimization Services Program? Request here: anyscale.com/asset/anyscale…

Most coding agents can write Python, but that does not mean they know how to deploy Ray workloads. They still miss GPU memory constraints, use stale APIs, and generate configs that break at runtime. We built Anyscale Agent Skills, so your agent can ask the right questions, validate the deployment, and generate production ready Ray configs from one prompt. Blog: na2.hubs.ly/H051dd50

Notion went from 3 Spark jobs per region to 1 Ray job on Anyscale. 80% cost reduction. 10x query latency improvement. Now powers all of Notion AI. Mickey Liu's full story in the Ray Day Seattle recap: na2.hubs.ly/H04TLGq0

Ray Day NYC lineup includes Robert Nishihara + AI platform leaders + Ray builders. Invite-only, limited capacity. Request invite: na2.hubs.ly/H04W_1Q0

Boston: Build distributed AI with Ray. Live, hands-on session with Anyscale engineers Omar Shorbaji & Alicia Chua. May 19 @ Convene, One Boston Place. Invite-only: na2.hubs.ly/H04TK440

Next stop: London. Join us for Ray Day London on June 2 at Convene 200 Aldersgate, St Paul for practitioner-led talks and a hands-on workshop focused on scaling AI systems in production. Spots are limited (invite required): anyscale.com/ray-day-london

Heading to Google Cloud Next? Come meet the Anyscale team at Booth #4807 and attend 2 must-see sessions on Agentic AI and next gen training on GKE. See the full agenda and register here: na2.hubs.ly/H04S6bL0

Ray Workshop is coming to Boston! Join us at Convene, One Boston Place on May 19 for hands-on, 200-level training to scale AI workloads with Ray (Core/Data/Train/Serve). Request invite: anyscale.com/ray-workshop-b…

Ray Day: Seattle recap is here. Notion: 80% cost cut on embeddings with Ray on Anyscale. Salesforce: 200K token docs summarized at P95 < 15s. Uber: 20% better GPU utilization with heterogeneous Ray clusters. Apple: foundation model training at scale. Full story: na2.hubs.ly/H04MTFW0