Brewit
Brewit is a conversational data analyst that writes SQL and recommends charts from plain language questions over a data catalog of the customer's database, with notebooks, dashboards, an evaluation harness and an embeddable chat agent.
Brewit lets people ask business questions in plain language and answers with SQL, charts and Notion style notebooks and dashboards. It connects to PostgreSQL, MySQL, BigQuery, Snowflake, SQL Server and Databricks, optionally over an SSH tunnel, plus CSV files and a Brewit hosted workbook database.
Answers draw on a data catalog of table and column descriptions and visibility settings, sampled distinct values and a query library of verified SQL, retrieved through a RAG framework that Brewit says handles hundreds of tables. An Evaluation feature runs sets of test questions against the database so the team can verify the generated SQL and answers, and Monitoring shows every conversation on a data source with feedback. The chat agent can be embedded in a company's own application for its customers, authenticated by JWT. Workspaces have Owner, Admin and Member roles.
Brewit runs on AWS and uses the OpenAI API (GPT-4), sending metadata rather than data. Hosting in a customer's own cloud is coming, by Brewit's account. Pricing is by contact for a Growth plan and an Enterprise plan with API access and white labeling.
Vendor details
Canonical URL
https://brewit.ai
Category
Data analyst agent
Company status
independent
Use cases & customers
Integrations
Brewit connects to PostgreSQL, MySQL, BigQuery, Snowflake, Microsoft SQL Server and Databricks (SSH tunnel supported), CSV and a hosted workbook database. The chat agent embeds with JWT authentication, and an API at api.brewit.ai comes with the Enterprise plan.
In practice
A marketing lead asks which campaigns drove the most qualified pipeline last quarter, and Brewit writes the SQL against the warehouse, returns a chart and lets her drill further in a notebook.
A data team documents tables and columns in the data catalog, saves verified queries to the Query Library and runs a set of test questions through Evaluation before opening Brewit to the wider company.
A software company embeds the Brewit chat agent in its product so its customers can ask questions of their own data, signing each session with a JWT.
Agentic Index coverage score
6.5 / 14 capabilities · 46%
| Integrations & Tool Calling | Partial |
|---|---|
|
Connections cover PostgreSQL, MySQL, BigQuery, Snowflake, SQL Server and Databricks (optionally over an SSH tunnel), plus CSV and a hosted workbook database, and Brewit writes and runs SQL against them. Every connector is a database. Brewit names no action in other systems and no custom tools. SourceBrewit, docs.brewit.ai Integrations pagesread 2026-09-28 |
|
| Workflow Orchestration | Partial |
|
The chat agent clarifies large requests, writes and runs SQL and recommends charts, and notebooks assemble reports and dashboards. The product sets this sequence. Brewit names no branching, conditions, multiple agents or flow that the customer configures. Sourcedocs.brewit.ai/docs/data-agent/best-practices/large-databasesread 2026-09-28 |
|
| Knowledge Grounding & RAG | Full |
|
A data catalog holds table and column descriptions, types and visibility settings, and a Distinct Values feature samples string values. Brewit also builds an automated semantic layer for business logic. It retrieves context through a RAG plus LLM framework that, by Brewit's account, handles hundreds of tables without token limits. Sourcedocs.brewit.ai/docs/data-agent/train/data-catalogread 2026-09-28 |
|
| Human Oversight & Guardrails | Partial |
|
Admins set table and column visibility in the data catalog, and only owners and admins manage data sources. The agent asks clarifying questions before large queries to limit cost. Brewit names no step where a person approves a query before it runs. Sourcedocs.brewit.ai/docs/workspace/access-controlread 2026-09-28 |
|
| Security, Identity & Governance | Partial |
|
Workspaces have Owner, Admin and Member roles, with a permissions table over chat, charts, data sources and workspace settings, plus table and column visibility. For audits and certifications Brewit points to AWS's and names none of its own. Data is encrypted in transit and at rest. Brewit does not say whether it supports SSO. Sourcedocs.brewit.ai/docs/others/securityread 2026-09-28 |
|
| Observability & Auditability | Partial |
|
Chat shows the SQL Brewit wrote and the tables it used. Monitoring lists all conversations on a data source with chat history and team feedback. Brewit names no audit log, step trace or export of what the agent ran. Sourcedocs.brewit.ai/docs/data-agent/monitorread 2026-09-28 |
|
| Memory & State Persistence | Partial |
|
Users save verified queries with titles and descriptions to a Query Library, which Brewit says keeps future answers accurate. People maintain this store for the agent, and it carries across sessions. Brewit names no memory that the agent writes for itself. Sourcedocs.brewit.ai/docs/data-agent/train/query-libraryread 2026-09-28 |
|
| Deployment & Data Residency | Not documented |
|
The service runs on AWS, and no hosting region is named or offered as a choice. Hosting Brewit in a company's own cloud is coming in the near future, by Brewit's account, so it is not available yet. Sourcedocs.brewit.ai/docs/others/securityread 2026-09-28 |
|
| Prebuilt Agents, Templates & Packs | Not documented |
|
Suggested questions help start an analysis. Brewit offers no prebuilt agents, templates or packs. SourceBrewit, docs.brewit.ai and changelogread 2026-09-28 |
|
| Triggers & Channel Coverage | Full |
|
The chat agent embeds in a company's own application, where that company's customers, authenticated by JWT, query their data by message. Brewit names no schedules or alerts. Its Slack is only a community workspace. Sourcedocs.brewit.ai/docs/data-agent/embedread 2026-09-28 |
|
| Model Flexibility & Routing | Not documented |
|
Brewit uses the OpenAI API (GPT-4) and sends it only metadata. It does not say whether customers can choose a model. Sourcedocs.brewit.ai/docs/others/securityread 2026-09-28 |
|
| APIs, SDKs & MCP Extensibility | Partial |
|
An API at api.brewit.ai/v1 takes workspace API keys, and API access comes with the Enterprise plan. Its only endpoint is a JWT endpoint for embedding the chat agent. There are no endpoints to query or manage the platform, and Brewit names no SDK or MCP server. Sourcedocs.brewit.ai/api-reference/introductionread 2026-09-28 |
|
| Testing, Debugging & Optimization | Full |
|
Evaluation lets the customer create test questions by hand or from a CSV, run them against the database and verify the SQL and answers Brewit generates. It serves as a test harness for the text to SQL agent. Brewit names no numeric score for those runs. Sourcedocs.brewit.ai/docs/data-agent/evaluationread 2026-09-28 |
|
| Browser & Computer Use | Not documented |
|
Brewit works through database connections and SQL. It names no browser, desktop or computer control. Sourcebrewit.airead 2026-09-28 |
|
The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Pricing
Contact sales (Growth and Enterprise plans, no prices published)
plan allowances (messages, data sources, workbooks); rates not published
What is public
The plan names and their allowances only; no prices, trial or metering are published.
Variable cost rationale
Growth caps messages at 2,000 a month; how usage past the cap is billed is not published.
Sales call required
Yes, required for paid access
Free / trial
Not published
Related vendors
- Actian AI Analyst — Wobby, now sold as Actian AI Analyst, answers plain language…
- AnswerRocket — Enterprise AI agent platform for analytics built around Max, which…
- AskEarth — Geospatial intelligence agent (GIA) that answers plain language…
- AskYourDatabase — AskYourDatabase is a conversational data analyst that turns plain…
- Athenic AI — Athenic AI is a data analyst agent that connects databases,…
- BlazeSQL — BlazeSQL is an AI data analyst that turns plain English into SQL…
Alternatives to Brewit
The closest documented capability profiles to Brewit among data analyst agents tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.
- BlazeSQL9.0 / 14Adds documented Deployment & Data Residency
- Actian AI Analyst9.5 / 14Adds documented Deployment & Data Residency and Prebuilt Agents, Templates & Packs, among others
- Athenic AI6.5 / 14Fuller documented coverage on Security, Identity & Governance and Observability & Auditability
- Defog.ai7.5 / 14Adds documented Deployment & Data Residency
- Lumi AI7.5 / 14Adds documented Deployment & Data Residency and Prebuilt Agents, Templates & Packs
- Pluvo6.5 / 14Adds documented Deployment & Data Residency and Prebuilt Agents, Templates & Packs
Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded