NVIDIA open-sources Auto Ontology: an auditable semantic layer for enterprise data
aigclink · x · 2026-11-04
NVIDIA has open-sourced auto-ontology on GitHub (Apache-2.0, 3.1k stars), a Python framework for building semantic, graph-based reasoning layers over enterprise data. It enables deterministic query generation and explainable answers for LLM and agent workflows.
It targets two pain points:
- LLMs can query tables but don't know business semantics: asking "last quarter's revenue in East China" fails because no such column exists — the model needs to know which fields compose revenue, which regions count as East China, whether to use order or invoice dates, and which tables join. Auto Ontology captures these definitions so queries follow certified logic.
- Analysts waste time re-explaining: questions like "why doesn't this revenue match the finance report" get固化 into a reusable, auditable semantic layer.
The workflow starts with Discover: scanning database metadata to auto-detect business entities, fields, and relationships. The repo ships an MCP interface, skills, a frontend, and Helm deployment configs for direct agent integration.
More from coding & agent
- Dev proposes LLM-generated ultra-rare content to build anime-style MMORPGs — rickasaurus · 2026-11-04
- Dev builds Chrome extension to bring back the classic Twitter bird on X — joonasvirtanen · 2026-11-04
- Developer lets his app record 'wishes' and auto-spawn OpenCode background agents to fulfill them — SubtleGradient · 2026-11-04
- Post-train your model on your own agent harness with OpenHands, Harbor, SkyRL and Snowflake — StasBekman · 2026-11-04
- Google Antigravity redesign leaked with context usage and quota indicators — clmt · 2026-11-04
- Microsoft's 4B Mage-VL-Cowork Hits 66% on OSWorld-Verified, Near Human Level — microsoft · 2026-11-04