In this episode, I’m talking with Vincent Warmerdam about treating LLMs as just another API in your Python app, with clear boundaries, small focused endpoints, and good monitoring. We’ll dig into patterns for wrapping these calls, caching and inspecting responses, and deciding where an LLM API actually earns its keep in your architecture.

Episode sponsors

Seer: AI Debugging, Code TALKPYTHON
NordStellar
Talk Python Courses

Links from the show

Vincent on X: @fishnets88
Vincent on Mastodon: @koaning

LLM Building Blocks for Python Co-urse: training.talkpython.fm
Top Talk Python Episodes of 2024: talkpython.fm
LLM Usage - Datasette: llm.datasette.io
DiskCache - Disk Backed Cache (Documentation): grantjenks.com
smartfunc - Turn docstrings into LLM-functions: github.com
Ollama: ollama.com
LM Studio - Local AI: lmstudio.ai
marimo - A Next-Generation Python Notebook: marimo.io
Pydantic: pydantic.dev
Instructor - Complex Schemas & Validation (Python): python.useinstructor.com
Diving into PydanticAI with marimo: youtube.com
Cline - AI Coding Agent: cline.bot
OpenRouter - The Unified Interface For LLMs: openrouter.ai
Leafcloud: leaf.cloud
OpenAI looks for its "Google Chrome" moment with new Atlas web browser: arstechnica.com

Watch this episode on YouTube: youtube.com
Episode #528 deep-dive: talkpython.fm/528
Episode transcripts: talkpython.fm

Theme Song: Developer Rap
🥁 Served in a Flask 🎸: talkpython.fm/flasksong

---== Don't be a stranger ==---
YouTube: youtube.com/@talkpython

Bluesky: @talkpython.fm
Mastodon: @talkpython@fosstodon.org
X.com: @talkpython

Michael on Bluesky: @mkennedy.codes
Michael on Mastodon: @mkennedy@fosstodon.org
Michael on X.com: @mkennedy

DFH- James Gunn's "Superman: Legacy"