Deepnote

What it does

Deepnote is a collaborative data workspace built around notebooks that combine Python, SQL, and no-code tools. It takes in data from connected warehouses, databases, and uploaded files such as CSVs, and produces notebook results, charts, dashboards, data apps, scheduled runs, and APIs that serve notebooks. Its built-in Deepnote Agent can generate, edit, and explain code, run blocks, and inspect outputs across a notebook in Edit mode.

Teams use it to explore data in shared projects, then publish results as dashboards or apps. The workspace also supports model training and serving, and autonomous data agents. External AI clients such as Claude, Cursor, VS Code, and Codex can connect through the Deepnote MCP server at deepnote.com/mcp to create notebooks, edit blocks, run notebooks, and manage integrations, within the permissions of the connected API key or OAuth grant.

Deepnote suits data analysts, data scientists, and data teams working in SQL and Python. Access starts with a sign-up, and a Free plan offers limited Agent requests, up to 3 editors and 5 projects, and basic machines. Free-plan AI and machine quotas are restricted, the Team plan is priced per editor, and Enterprise features are offered on a custom basis.

What it can do

• Write and run SQL and Python blocks in shared notebooks
• Use Deepnote Agent to generate, edit, explain code and run blocks
• Build autonomous data agents and train and serve models
• Schedule notebook runs and deploy notebooks as APIs
• Publish notebooks as dashboards or data apps with input blocks
• Choose machine types including GPU options
• Expose the workspace to AI clients via the Deepnote MCP server with OAuth or API key

Works with

• PostgreSQL
• MySQL
• BigQuery
• Snowflake
• Redshift
• Databricks
• MongoDB
• Google Drive
• GitHub
• GitLab

Before you choose

• Deepnote Agent uses third-party model providers such as Anthropic and OpenAI; review the data privacy docs before enabling block-output access.
• MCP tools can modify workspace content, and deleting files cannot be undone; use read-only access where appropriate.
• Free plan machines and AI requests are limited, and Team plan AI and compute credits are capped monthly.

Listed under

• Coding
• MCP and agent tools
• AI agent
• Collaboration
• Dashboards
• Data analytics
• Data notebook
• Machine learning

Get started

A Free plan is offered at sign-up.

1. Create an account at deepnote.com/sign-up and open a project.
2. Connect a data source under integrations, such as BigQuery or Snowflake.
3. Optionally create a workspace API key under Settings & members > Security > API keys and add the MCP endpoint to an AI client.

Team shows '$39 per editor/month billed yearly' with a 20% yearly discount marker and a 14-day free trial. Enterprise is custom contract. Free is listed as forever free with limits.

14-day free trial.

Source: deepnote.com/pricing, checked 30 Oct 2026. Prices can change; check before you buy.

Sources for this overview

Product pages, documentation, and pricing checked 30 Oct 2026. Each source supports part of the overview; availability and prices can change.

• Welcome to Deepnote - Deepnote docs ↗deepnote.com
• Deepnote Agent - Deepnote docs ↗deepnote.com
• Deepnote Pricing ↗deepnote.com
• Deepnote MCP - Deepnote docs ↗deepnote.com
• Deepnote: Collaborative analytics & data science notebook ↗deepnote.com

Company website

deepnote.com: product, pricing and policy pages

Hacker News

Launch posts and discussion

YC directory mirror

Batch and company name

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

manual_review

Last checked 30 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.