NVIDIA Proposes NOOA: Abstracting AI Agents as Python Objects
omarsar0 · x · 2026-07-30
NVIDIA suggested a new approach to building AI agents called NOOA, which models agents directly as Python objects.
This replaces the current scattered approach of prompt templates, tool schemas, callback code, and workflow graphs with a single abstraction:
- Methods: Act as actions the model can take
- Fields: Hold state
- Docstrings: Serve as prompts
- Type annotations: Act as contracts
This convention places the boundary between probabilistic (LLM-completed at runtime) and deterministic (normal Python) behavior directly in the source code, making agent behavior testable, traceable, and refactorable with existing codebase tools.
Related event: NVIDIA Open-Sources Object-Oriented AI Agent Framework NOOA(3 posts)→
More from coding & agent
- Reflection's Beam, a 501B-parameter open MoE model with 1M context, lands on Hugging Face under Apache 2.0 — QGallouedec · 2026-10-16
- 'LLM code is sloppy' rebutted with real ffmpeg code: identical if/else branches in a top library — kingslyj · 2026-10-16
- Office-editing tool calls need file, worktree and unit: lessons from dsh-univer-office — FunDistribution6537 · 2026-10-16
- Architect MD workflow: Gemini one-shot outputs that beat Opus, via 5-lens plan critique — oikk01 · 2026-10-16
- Prove a coding agent actually ran the tests: check artifacts, not approvals — ComprehensiveShake76 · 2026-10-16
- Building an acceptance harness for AI spreadsheet edits: allowlists, recalc checks, idempotency — ukdicha-modakkk · 2026-10-16