Dot

About Dot

Dot is an AI data analyst that answers business questions in the channels a team already works in — Slack, Microsoft Teams and email — rather than in yet another BI tool. Someone asks in plain English, Dot finds the right tables, writes the SQL, runs it and returns an answer with a chart, so the long tail of ad-hoc requests that normally queues behind a data team gets served immediately. Beyond one-shot questions it runs deep analyses with drill-downs and a stated methodology, and generates recurring executive reports as PowerPoint decks on a schedule, built from live data. The piece that separates it from a generic text-to-SQL wrapper is the context agent: it pulls metric definitions and documentation from the systems that already hold them — the vendor cites Tableau dashboards, Snowflake query history and Confluence data dictionaries — cross-checks definitions against how metrics are actually queried in practice, and writes the documentation that is missing, all under governance. You can also train it directly with instructions, examples and business rules. The company benchmarks against DABStep, the 450-plus-task multi-step financial analysis benchmark published by Adyen and Hugging Face, and publishes its scores against a human analyst baseline rather than only against other models.

Key Features

Tags