Keynote: Building an Open Agentic Future
summary
Pydantic AI is an open agent framework that claims fast growth and a practical alternative to closed lab platforms, with concrete choices: switch models in one line of code (OpenAI GPT‑5 example), emit OpenTelemetry to Logfire or any backend for observability, use MCP servers for context and tool access, and extract structured outputs with Pydantic type hints. The talk argues major labs are centralizing value while Pydantic builds engineering-first tooling that remains portable and replaceable against firms with trillions in market cap.
AI-written summary · about
chapters (6)
- 0:00Introduction and thesis
- 0:40The vertical integration trap
- 4:13Open standards as an alternative
- 7:07Meritocracy vs. monopolies
- 8:36Lessons from the past
- 12:29Practical implementation
related talks
05mKeynote: How Open Protocols Unleash AI’s Next Wave of Creativity and Impact
30mKeynote: Fireside Chat
10mKeynote: The Open Source Runway: Building the Foundation for the Agentic Era
01h 04mBuilding AI Applications the Pydantic Way
16mKeynote: Open Source for AI and AI for Open Source
19mKeynote: Towards Building Safe and Secure Agentic AI
17mKeynote: An OPEN Path to Super Intelligence
42mKeynote: Does Open Data Science Lead to the End of the World?