OpenAI case study: V7 Go gives enterprise agents long-term memory via Context Graph
xiaohu · x · 2026-09-22
OpenAI published a case study on V7 Go, the enterprise agent platform from European startup V7, tackling why enterprise agents behave like first-day hires: they don't know which document is current, whether entities across systems are the same, or which source backs a conclusion.
The approach: pre-extract entities, relations, facts and citations into a source-referenced Context Graph, falling back to document-level RAG when the graph is insufficient; recent conversations stay in context while older material is retrieved from the graph. Execution is tiered by difficulty—Luna for high-throughput extraction, Terra/Sol for complex tool calls, Astra for hard graph queries—with reported improvements in cost, error rate and complex-query accuracy.
The architecture fits businesses dense in cross-document relations where every conclusion must trace to evidence; simple Q&A is better served by plain RAG.
More from coding & agent
- Dev recommends logging every endpoint request to spot brute-force attackers — Kyrannio · 2026-09-22
- Debate: will all mainstream software be formally verified by 2030? Legacy debt says no — luislamb · 2026-09-22
- OpenMuse: open-source AI coworker that browses the web in a disposable Linux desktop — aniketmaurya · 2026-09-22
- Turn detection experiment: Jev hits 16/16 while tiny-BERT Laya flunks on own data — tobowers · 2026-09-22
- What engineers check before adding a new LLM provider — Rama_Surasani_ · 2026-09-22
- Builder: AI productivity gains in software are 'truly unbelievable' — omnivaughn · 2026-09-22