Orchestrating Python Functions Natively in Argo Using Hera
summary
Hera compiles Python workflows into Argo YAML, treating every decorated function as a reusable template so you can unit test templates and still run them on Kubernetes. Script templates can be inlined or built into a container image via a script Runner so inputs are deserialized, type-checked (pydantic integration) and function return values flow back into the workflow. Bloomberg rewrites Hera for a major v5 release and enables 40 teams to adopt Argo by using Python-functions-as-templates, cookiecutter, testing and docs generation.
AI-written summary · about
chapters (10)
- 0:00Intro
- 0:50Why Argo Workflows
- 4:22Lets Code
- 5:51Script Templates
- 7:22Hera Runner
- 8:51Pantic
- 10:11Testing
- 10:59Bloomberg
- 15:01Use Cases
- 17:34Conclusion
related talks
25mData Science Workflows Made Easy: Python-Powered Argo for Your Organization
27mCreating extensible workflows with off-label use of Python
44mImplementing a Workflow Engine in Python
26mIn house to open source Stitching the past to the futu...
23mGitHub Actions & Python
31mHamilton:Natively bringing SWE best practices to python data transformations
31mPython + Github Actions + Kubernetes : Your Path to achieve Nirvana
32mDeploy Django: GitOps & Kubernetes Made Easy