Lightning Talk: Automating Map-building Pipelines for Safe Autonomous Driving...
summary
TuSimple automates its map-building pipelines using Argo Workflows on Kubernetes to get open-source introspection, high availability and horizontal scalability, and human-in-the-loop controls that pause pipelines and wait for manual decisions. Pipelines include backprocessing, patching, CI/CD and ML workflows, and engineers add event aggregation customization and a retry an arbitrary node feature to avoid rerunning whole runs. The system runs in production and the Argo project lists 800 contributors and 200+ end users.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 0:33Why Argo Workflow
- 1:58Collaborations
- 2:59Workflows
- 3:42Use Cases
- 4:26Community Contributions
- 5:19Adoption
- 6:40Outro
related talks
24mBuilding an AI-Powered, Paved Road Platform with Cloud-Native OSS
25mEvolving Argo Workflows for AI: New Features, Tuning, and Multi-Tenant Isolation
22mBuilding AutoML Pipelines With Argo Workflows and Katib
54mScaling Machine Learning Pipelines in Cloud
08mLightning Talk: Enabling Hybrid Cloud Infrastructure with Argo Workflows
34mDevOps at LLM Speed: Using an AI Copilot for Kubernetes and Jenkins
27mEmpowering Autonomy: BYD's Journey Taming Million-Task Scale With Argo...
06mLightning Talk: Harnessing Generative AI Without Making a Faux Pas