Automating h2oGPTe Workflows with the Python API
- Generative AI
- Enterprise h2oGPTe
This lesson walks through productionizing an h2oGPTe workflow using its Python API instead of repeating the same manual chat every day. It shows a Jupyter notebook that connects to h2oGPTe with an API key, lists available LLMs, and merges a set of questions and answers into a single text file. It then creates a collection through the API, uploads the file to it, and starts a chat session that asks the same question used earlier in the manual interface, producing a Python list of results that later stages of the course use to build a Wave app for business users.
Key takeaways
- Why moving a repeated manual chat task into the Python API saves time for recurring workflows
- How to create a collection and upload a file to it programmatically using the h2oGPTe API
- How starting a chat session through the API returns a Python list that other code can use downstream
- What the Python API workflow is used for next: powering a Wave app for non-technical business users
Start owning your AI