Mirendil has announced a $1 billion Series A to expand its work on automating AI research and development. The November 12 announcement names Kleiner Perkins and Clearlake Capital as lead investors, with Google among the participants.
The funding backs a workflow in which researchers direct agents that handle implementation and experiments. Its practical expression is Lor, a platform Mirendil says it uses internally. Lor remains in private beta, so the investment should not be read as a general product launch.
In its product description, Mirendil says Lor can distribute work across hundreds of agents, provision compute and keep tasks running for days. People set the research direction and decide which results to trust.
The company also cites its own model-building results as evidence for the approach. Those statements are vendor claims; the financing announcement does not independently validate model performance or research productivity.
For teams considering this approach, access controls, compute budgets and reproducible experiment records become practical concerns as more work moves into long-running agent sessions.
