Factory Acquires Typedef to Give Its Droid Coding Agents Deeper Data Context
Typedef maps how data models, pipelines and dashboards depend on each other, so Factory's Droids can plan migrations and debug pipelines with clearer lineage.
Published
Last updated / Fact-checked on
Factory, the startup behind the enterprise coding agent Droid, announced on October 30 that it is acquiring Typedef. Typedef builds a context layer that maps how parts of an enterprise data system depend on one another. The deal came a day after Factory published Droid 1.0, a new engine for its agent. Terms were not disclosed.
Image: Illustrative photo of hands typing on a laptop keyboard. Fabian Irsara / Unsplash via Wikimedia Commons — CC0 1.0
What Typedef does
According to Factory's announcement, a change to a single data model can "ripple through definitions, pipelines, and dashboards." Without a clear view of those dependencies, Factory says, it is harder to plan a migration, check a refactor or explain the result.
Typedef builds a context layer from code that makes those relationships visible. Factory supplies the harness that lets Droids work inside a customer's environment. Together, the company says, the two will give engineers "clearer lineage and evidence as Droids plan migrations, debug pipelines, and refactor models."
For example, renaming a field in a database model may require matching changes in a transformation pipeline, a metric definition and a dashboard. An agent that sees only the code it is editing can finish its task and still leave those downstream pieces broken. Lineage data is meant to surface them before a change ships.
Factory did not disclose the price, financial terms or an integration timeline. It also has not published results showing how much Typedef improves Droid's accuracy.
The founders
Typedef co-founders Yoni Michael and Kostas Pardalis are joining Factory. According to the announcement:
- Yoni Michael co-founded Coolan, which used machine learning and infrastructure analytics to help enterprises improve data center reliability and efficiency. Coolan's technology ran on more than 100,000 servers before Salesforce acquired it in 2016. He later led infrastructure engineering teams at Salesforce and Tecton.
- Kostas Pardalis founded Blendo, which more than 400 companies used, and later led product at RudderStack and Starburst.
Factory says the two teams share "a technical conviction: customers should control their infrastructure."
Factory's run of deals
Factory was founded in 2023 by Matan Grinberg and Eno Reyes and is based in San Francisco. It describes Droids as agents that "plan, build, validate, and improve work" inside each company's infrastructure, policies and approved models.
The Typedef deal follows a busy few weeks:
| Date | Announcement |
|---|---|
| Sept. 15 | Raised $200 million at a $5 billion valuation, bringing total funding to over $400 million |
| Oct. 9 | Acquired Mentlio, a Y Combinator startup building in-harness model routing and context compression to cut AI token spend |
| Oct. 19 | Partnership with Databricks to connect Droids with data and models governed in Databricks; Factory is sold through Databricks Marketplace |
| Oct. 29 | Published Droid 1.0, a new engine for its enterprise agent |
| Oct. 30 | Acquiring Typedef |
The Databricks partnership points in the same direction as Typedef. In its announcement, Factory said Droids can trace "how an application depends on a stored procedure, which data pipeline creates its inputs, and where a change could affect downstream behavior." Typedef adds a dedicated layer for that kind of mapping.
Why it matters
For enterprise teams, generating correct code is only part of the job. The harder part is changing an interconnected system without breaking something two steps away. Factory's recent deals point to a bet that coding agents need a map of the systems they change, and an accounting of what they cost, as much as better models.
The open question is how well that works in practice. Factory has published neither a timeline for building Typedef into Droids nor benchmarks for the combined product.
What enterprise teams can do now:
- If you run Droids against data platforms, ask Factory when Typedef lineage will reach your deployment.
- Inventory which pipelines and dashboards depend on the models you plan to migrate, whatever tool you use.
- Ask for evidence, such as dependency reports or test results, with any agent-made change to data models.
More coverage is on our Artificial Intelligence hub.
How we reported this
We verified the deal, founder biographies and company background against Factory's October 30 announcement. Funding came from Factory's September 15 post. The Databricks and Mentlio details came from Factory's own posts, with the Databricks deal confirmed in a Business Wire release. Droid 1.0's date comes from Factory's news page. We corrected an earlier description of Mentlio and removed tracking parameters from links.
Published and fact-checked October 30, 2026.
Sources
- Factory, "Factory Acquires Typedef: Bringing Data Context to the Software Factory," Oct. 30, 2026: https://factory.com/news/typedef
- Factory, "Factory raises $200M at $5B valuation," Sept. 15, 2026: https://factory.com/news/5-billion-valuation
- Factory, "Factory Acquires Mentlio: Building Intelligent Token Allocation for the Software Factory," Oct. 9, 2026: https://factory.com/news/mentlio
- Factory, "Factory and Databricks: Building Software Powered by Data," Oct. 19, 2026: https://factory.com/news/databricks-partnership
- Business Wire via Yahoo Finance, "Factory Partners with Databricks to Bring Autonomous Software Development to the Enterprise," Oct. 19, 2026: https://finance.yahoo.com/technology/ai/articles/factory-partners-databricks-bring-autonomous-130000881.html
- Factory, News: https://factory.com/news
DailyMech used AI tools to help research and draft this article; it was fact-checked by our editors.