Devin adds managed Devins for parallel VM task execution
Cognition updated Devin so one session can break down large work and delegate subtasks to worker Devins running in separate VMs. It matters for audits, migrations, and QA runs where one long-context agent is slower than explicit parallelism.
TL;DR
- Cognition shipped a new Devin mode where one Devin session can "break down large tasks" and delegate them to parallel worker Devins, with each worker running in its own VM launch thread.
- The company’s announcement post says the manager Devin handles decomposition, assignment, progress tracking, and result synthesis, making the feature aimed at multi-step engineering work rather than single-ticket code edits.
- Cognition’s use-case examples specifically frames the launch around parallel QA, code migrations, security and dependency audits, and refactors—jobs where explicit task splitting can beat one agent carrying all context alone.
- Devin is also now exposed through Poke, with Cognition’s Poke post pitching a text-driven way to start sessions and track progress from that collaboration surface.
What shipped?
The core change is orchestration. According to Cognition’s launch thread, Devin can now "manage a team of Devins," spawning parallel workers that each run in their own VM instead of forcing one long-running session to do everything serially. The product claim is that Devin will also improve "over time" at breaking work down for a given codebase launch thread.
Cognition’s announcement post adds the implementation detail missing from the tweet: the top-level Devin acts as coordinator, assigning subtasks, monitoring progress, resolving conflicts, and compiling results. The same post says users can message individual workers, schedule follow-ups, pause or terminate sessions, and watch compute usage, which makes this closer to managed multi-agent execution than a simple fan-out job runner announcement.
Where does this matter in practice?
The launch material is most concrete around workflows that already decompose cleanly. Cognition’s use-case examples calls out parallel QA across web pages, large-scale migrations, security and dependency audits, and codebase refactors. Those are the kinds of tasks where isolated environments matter, because each worker can run commands, tests, and verification independently before handing results back to the coordinating session.
The same day, Cognition said Devin is now available in Poke Poke post. Poke’s linked product page describes a text-first interface for starting Devin sessions, monitoring progress, and collaborating around the work Poke page. That does not add new agent capability by itself, but it does broaden where teams can invoke the new managed-Devins workflow.