# Timbal

> Timbal is the production AI platform enterprise teams use to build, deploy, and govern agents, workflows, and knowledge bases on the models they choose. The platform is open at the framework layer (Python SDK), governed at the enterprise layer (sovereign hosting, owned model keys, access controls, run-level observability), and developer-first at every layer in between.

Built and operated by Timbal Tech S.L. in Barcelona.

## Product

- [Studio overview](https://timbal.ai/studio): The integrated development surface where teams build agents, workflows, knowledge bases, and the interfaces that ship them.
- [Agents](https://timbal.ai/studio/agents): Production AI agents with typed tools, hybrid knowledge search, eval gating, and full observability on every run.
- [Workflows](https://timbal.ai/studio/workflows): Python-first typed workflow graphs that orchestrate agents, tools, and APIs with scheduling, streaming, retries, and trace replay.
- [Interfaces](https://timbal.ai/studio/interfaces): Ship the same agent runtime to chat, email, voice, and product UIs with one auth layer and one audit log across every channel.
- [Knowledge bases](https://timbal.ai/studio/knowledge-bases): Combine structured tables and files in one queryable layer. Define schemas in plain language, query with SQL, return answers with citations.
- [Integrations](https://timbal.ai/studio/integrations): Native integrations, custom tools, and MCP servers so agents reach the systems your team already uses.

## Developers

- [Developers hub](https://timbal.ai/developers): Open-source Python stack — typed IO, stream-native execution, provider fallbacks, built-in evals.
- [Framework](https://timbal.ai/developers/framework): Python primitives for defining agents and workflows in code with reproducible behavior.
- [SDK](https://timbal.ai/developers/sdk): TypeScript SDK to query knowledge bases and run workflows from Node.js, browsers, and edge runtimes.
- [CLI](https://timbal.ai/developers/cli): Manage projects, environments, and deployments from the terminal.
- [API reference](https://timbal.ai/developers/api): HTTP API for connecting enterprise systems and developer tools to Timbal workloads.
- [MCP](https://timbal.ai/developers/mcp): Model Context Protocol surface so external clients (Claude, ChatGPT, Cursor) can call Timbal tools and knowledge.

## Enterprise

- [Enterprise overview](https://timbal.ai/enterprise): Sovereign hosting, owned model keys, access controls, and observability governed from day one.
- [ACE](https://timbal.ai/enterprise/ace): Evaluation, governance, and measurable operational outcomes for AI applications.
- [Security](https://timbal.ai/enterprise/security): Permissions, auditability, deployment control, and the platform security posture.
- [Trust center](https://trust.timbal.ai/): Compliance certifications, security controls, and data-handling documentation.
- [Use cases](https://timbal.ai/enterprise/use-cases): Sector-specific AI workflow and agent patterns enterprise teams deploy in production.
- [Case studies](https://timbal.ai/enterprise/case-studies): Customer outcomes with named protagonists, scoped problem statements, and shipped results.

## Resources

- [Blog](https://timbal.ai/blog): Product updates, technical deep dives, and engineering field notes from the Timbal team.
- [Playbook](https://timbal.ai/playbook): Field-tested patterns for shipping AI agents, workflows, evals, and knowledge bases into production.
- [Changelog](https://timbal.ai/changelog): Every Timbal release in one feed.
- [Documentation](https://docs.timbal.ai): Product guides, API references, and how-tos with AI search across every page.

## Company

- [Pricing](https://timbal.ai/pricing): Free forever for individuals, paid plans from €25 a month, custom for enterprise. Same full stack on every plan.
- [Demo](https://timbal.ai/demo): Schedule a guided walkthrough of agents, workflows, integrations, and enterprise controls.
- [Contact](https://timbal.ai/contact): Get in touch with the Timbal team.
- [Partners](https://timbal.ai/partners): Partner with Timbal to deliver production AI for enterprise customers.
- [Careers](https://timbal.ai/careers): Open roles in Barcelona and remote across the EU.

## Legal

- [Privacy policy](https://timbal.ai/legal/privacy-policy)
- [Terms](https://timbal.ai/legal/terms)
- [Cookie policy](https://timbal.ai/legal/cookie-policy)
- [Data Processing Addendum](https://timbal.ai/legal/dpa)
- [Sub-processors](https://timbal.ai/legal/subprocessors)