Two cell-therapy factory vendors stitch their robots and software into one floor

Autolomous and Cellular Origins have completed the first end-to-end integration of their cell-therapy manufacturing platforms at a UK government-backed testbed, the companies said this month. The two vendors connected a software layer, designed to knit devices, data, and AI across a factory, with a robotics platform built to handle per-patient cell-therapy steps. The combination points at cell therapy manufacturing's real bottleneck: not biology, not robots, but the dozen disconnected software and hardware stacks that have to talk to each other for every patient batch.

The integration is being demonstrated at the Cell and Gene Therapy Catapult's Digital and Automation Testbeds, a UK facility funded through Innovate UK that exists specifically to let vendors prove integration claims before they reach a real GMP (good manufacturing practice) production floor. Autolomous' autoloMATE software links devices, data, and AI across a manufacturing process and across the supply chain, with intellectual-property safeguards so different process owners, including competing therapy developers who share a contract manufacturer, can share data without exposing proprietary steps. Cellular Origins' Constellation, launched in May 2023, is a configurable robotics system that uses mobile robots and sterile fluid transfer to handle cell-therapy steps, positioned as a modular alternative to the bespoke, largely manual workflows that still dominate advanced therapy manufacturing.

Cell therapy manufacturing is fragmented in a way most pharmaceutical production is not. A single patient batch can move through multiple vendors' instruments, each with its own data silo, and end-to-end visibility is limited to whatever a human can stitch together from logs and batch records. The reason scaling is so hard: cell therapies are made one patient at a time, often by contract development and manufacturing organizations (CDMOs) running hand-built processes, so the factory problem arrives before the science problem does.

Alexander Seyf, the CEO of Autolomous, calls the bottleneck delivery infrastructure, arguing that getting approved cell therapies to patients is constrained by manufacturing capacity rather than by the underlying science. Edwin Stone, the CEO of Cellular Origins, calls scalability a system-evolution problem in which the right answer is not "add another robot" but a rethought orchestration layer across the whole workflow.

Cellular Origins hit a first robotic-automation milestone with Fresenius Kabi, an industrial-scale bioprocessing vendor, in January 2026, on a separate workflow. The Autolomous pairing extends that integration story into the data and AI layer, and the testbed deployment positions the companies to court CDMOs and therapy developers who want orchestration without ripping out their existing instruments.

Neither CEO's claim is independently verified by this announcement. A vendor partnership release with two CEOs on the record, a testbed integration, and a UK-government-funded proof site is not the same as a validated factory win. The release does not name a commercial deployment, a cell-therapy customer running the combined stack in production, or any third-party throughput, cost-of-goods, or comparability benchmark. Trade-press pickups of the announcement repeat the same vendor-supplied claims without independent reporting.

An independent cell-therapy manufacturer who runs a batch through the combined stack would surface the answers the press release does not: how the IP safeguards hold up when more than one process owner is involved, whether the data layer reduces batch failures or only adds another dashboard, and what the integration costs against a hand-built workflow. Until that lands, the headline is the testbed integration itself: two vendors proving their stacks can talk, while the field's fragmentation problem waits for a real customer reference.