---
schema_version: "newruntime-agent-readable-v0.1"
type: "raw_signal"
id: "tg-2485"
slug: "openai-built-a-data-agent-across-ninety-thousand-tables"
title: "OpenAI Built a Data Agent Across Ninety Thousand Tables"
description: "A large internal data environment requires metadata discovery, permission-aware retrieval, query validation, and feedback loops beyond a generic text-to-SQL prompt."
observed_at: "2026-06-23"
why_it_matters: "This dated record adds public evidence to the company memory needs write loops, skills become portable capability layer analysis and keeps the claim auditable as the underlying products and practices change."
novelty: "notable"
verification_level: "source-linked"
signal_type: "field-report"
evidence_kind: "independent-analysis"
status: "published"
telegram_message_id: 2485
telegram_url: "https://t.me/qwgai/2485"
topics: ["data-agents","enterprise-data","verification"]
entities: []
related_patterns: ["company-memory-needs-write-loops","skills-become-portable-capability-layer"]
source_urls: ["https://blog.bytebytego.com/p/how-openai-built-its-data-agent"]
import_batch: "telegram-2026-07-17-new-sources-v3"
routes: {"html":"https://newruntime.com/signals/openai-built-a-data-agent-across-ninety-thousand-tables/","markdown":"https://newruntime.com/signals/openai-built-a-data-agent-across-ninety-thousand-tables.md","json":"https://newruntime.com/signals/openai-built-a-data-agent-across-ninety-thousand-tables.json"}
source_format: "telegram-export-normalized-json"
---

# OpenAI Built a Data Agent Across Ninety Thousand Tables

## Observation

A large internal data environment requires metadata discovery, permission-aware retrieval, query validation, and feedback loops beyond a generic text-to-SQL prompt.

## Why it matters

This dated record adds public evidence to the company memory needs write loops, skills become portable capability layer analysis and keeps the claim auditable as the underlying products and practices change.

## Provenance

Normalized from QWG AI Telegram message 2485. The original Russian-language record remains available at https://t.me/qwgai/2485.