Agility Robotics Opens 60,000-Square-Foot Fremont Training Site Near Tesla’s Planned Optimus Production

The Bay Area expansion adds robot-training capacity for Digit as Agility pursues warehouse and manufacturing work already generating revenue, while Tesla prepares its own humanoid manufacturing effort nearby.

Agility Robotics is opening a 60,000-square-foot facility in Fremont, California, to train its Digit humanoid robots, placing the company near the Tesla factory where Optimus manufacturing is expected to begin this year.

The new site expands Agility’s presence in a region that is becoming a concentrated battleground for humanoid robotics. Tesla has made Optimus a major strategic priority, though its expected manufacturing start has not been confirmed and the company has not established a commercial deployment timeline for external customers.

Agility, by contrast, is already operating Digit in warehouse and manufacturing settings. The company says its bipedal robots carry totes and bins for customers including Amazon, GXO, Schaeffler, and Toyota Motor Manufacturing Canada. Agility says it has secured $300 million in contract orders for its robots.

The company has not disclosed how many Digit units it has built or deployed. Independent observers have estimated that dozens of robots have participated in pilot or revenue-generating deployments. Agility has said that Digit moved 100,000 totes at a GXO logistics facility.

For operators, the Fremont facility matters less as a real estate move than as an indication that Agility is investing in the practical work of preparing robots for repeatable deployment. Training humanoids for commercial facilities requires more than locomotion and object handling. The systems need to operate around people, connect with warehouse-management and IT systems, and meet site-specific safety, regulatory, and compliance requirements.

“We have commercialized,” Agility CEO Peggy Johnson said. “We now know what it takes to walk into these facilities and meet their safety bars, their regulatory bars, compliance, plug into their IT infrastructure, plug into their warehouse management system.”

Johnson welcomed Tesla’s proximity, arguing that the presence of another major humanoid effort could help validate the market. “It’s great to have Tesla in the same area as us, because really, for a long time Agility was out there alone, and it’s good to have others in the humanoid space,” she said.

Tesla’s Optimus program carries a substantially different scale of ambition. Elon Musk has said he expects Optimus to become useful outside Tesla sometime next year and has described it as potentially Tesla’s biggest product. Those targets remain forward-looking. Tesla’s planned manufacturing start and the timing of external deployment are not confirmed in the available details.

Agility’s immediate advantage is not necessarily robot volume. It is the accumulated experience of putting Digit into real industrial workflows. A humanoid robot must do more than demonstrate a task in a lab. It must recover from interruptions, work within a facility’s operating rules, avoid unsafe actions, and deliver enough uptime to justify integration work.

Agility co-founder and chairman Damion Shelton has described a split architecture for that problem. The company sees generative AI as a way to scale application development, but not as the control layer for safety-critical behavior.

“The analog with humanoids is all the safety stuff needs to go through a path that’s not generative AI,” Shelton said. “You don’t want to get creative with your safety stack.”

That distinction is central to the near-term commercial humanoid market. Generative models may help operators specify new tasks and reduce programming time, but warehouse buyers will judge systems on safety, integration effort, reliability, throughput, and cost per completed move.

The Fremont facility’s exact opening date and its full technical capabilities have not been disclosed. Agility has said it will use the site to train humanoid robots, but has not detailed the number of robots it will support, the types of training environments it will contain, or whether it will serve customer-specific validation programs.

The location nevertheless puts Agility closer to Tesla as both companies prepare for a more demanding phase of humanoid development. Tesla must translate Optimus from an internal manufacturing project into a product that can work outside its own facilities. Agility must turn early commercial deployments and contract orders into a repeatable, scalable operating model.

Agility is also pursuing a reverse merger that it expects will make it the first pure-play humanoid robotics company on public markets later this year. The Fremont expansion gives the company another operational asset as it makes that case to investors and customers: Digit is not only being developed as a general-purpose humanoid, but trained for the warehouse and manufacturing tasks that companies are willing to pay for now.