From Adoption to Production at Scale
summary
OpenAI moves a high‑throughput image pipeline onto Temporal Cloud, routing 1 billion images per week and growing 60x in a year by building a platform layer: an SDK wrapper, a config generator called gen temporal manage worker that produces worker specs and queues, a proxy in Go that enforces auth, routing and payload encryption, and a temporal operator that reconciles namespace, RBAC and K8s resources; 700+ namespaces, ~100 workers, and a four‑person team operate it.
AI-written summary · about
chapters (8)
- 0:00Introduction and speaker welcome
- 1:04Use cases for Temporal at OpenAI
- 4:38Platform journey and adoption strategy
- 10:39Paved road architecture details
- 16:16Production readiness and observability
- 23:00Outcomes and key lessons learned
- 24:05Q&A session begins
- 42:05Final Q&A and closing
related talks
33mFrom Prototype to Production: Making Agentic AI Systems Reliable
42mUsing Temporal Workflows for Seamless Dev + Research Collaboration & Continuous Algo D
01h 03mGoing production scale with Gen AI - Patterns for your API
28mBuild and scale AI apps with Azure OpenAI, AI Search, and HCP Terraform
51mCloud AI: How to Get Started Injecting AI Into Your Applications
17mAI TechOps: Data and Infra for AI Foundation to Scale and Manage 1000s of Cohesive Agents
19mNo Token Budget: Agentic Software Factories on Kubernetes
37mAccelerating the Journey of an AI Algorithm to Production with OpenFaaS