Amazon’s next phase of warehouse automation is less about unveiling a single new robot and more about connecting many existing systems into larger, more capable fulfillment centers. At its Delivering the Future event in Nashville, the company focused on AI in shopping, computer vision for delivery speed, and the way its robotics stack is being expanded inside real operations.
The clearest example is Shreveport, Louisiana, where Amazon is expanding a fulfillment center around a much larger version of Sequoia, its automated storage and retrieval system. The project shows how the company wants to increase robot density without always starting from a blank site.
Amazon’s robot fleet is broader than one headline number
Amazon says it has 750,000+ robots deployed in its U.S. fulfillment centers. That public figure is unchanged from 2023, but it does not describe the full range of machines working across the company’s operations.
The 750,000+ figure refers to autonomous mobile robots, or AMRs. These are the wheeled systems that move across warehouse floors, a category tied to Amazon’s 2012 acquisition of Kiva Systems. Amazon also calls these tote-moving machines its drive-train systems.
That AMR group includes several models, among them Proteus, the autonomous system unveiled at Re:Mars in 2022. Wheels still matter because Amazon’s fulfillment centers have many smooth concrete floors, which make mobile robots practical for a large share of internal movement.
But Amazon Robotics is no longer only about wheeled machines. Robotic arms have become the next major category. Robin, Cardinal, and Sparrow are each used for sorting and stacking objects, extending automation from movement into more complex handling tasks.
Sequoia points to a bigger warehouse architecture
The newest member of Amazon’s robotics family described in the source is Sequoia, introduced at the 2023 Delivering the Future event. It is an automated storage and retrieval system, similar in principle to systems offered by companies like AutoStore.
The first Sequoia system went online in 2023 at a Houston-area fulfillment center. Amazon has now announced a system 5x its size in Shreveport, Louisiana. That system sits at the center of a fulfillment center being expanded to more than 3 million square feet.
Amazon could build new robotic fulfillment centers from the ground up. Instead, the company is putting major emphasis on retrofitting existing brownfield warehouses. That approach uses existing sites, but it also requires Amazon to keep working around live delivery operations.
Tye Brady, Amazon Robotics chief technologist, described that challenge as “fix the airplane while it’s flying.” The phrase captures the operational difficulty: the company is adding more automation inside facilities that cannot simply stop functioning while new systems are installed.
Shreveport becomes the first “Gen 12” building
The Shreveport center is Amazon’s first “Gen 12” building. According to the company, it will ultimately use 10x the number of robots as previous facilities, though Amazon has not disclosed the exact number.
The scale is also visible in the physical footprint. Amazon says the site will cover 55 football fields' worth of fulfillment operations once it is fully running. The company says the facility will employ 2,500 humans.
More robots also mean more jobs focused on keeping the robotic systems operating. Brady said the new systems will come with 25% more RME, or reliability maintenance engineering, roles than before.
That detail matters because Amazon’s automation story is not framed only as replacing manual labor with machines. The company’s position is that robotics shifts more work toward areas where humans still have an advantage, especially when tasks require judgment rather than repetition.
When asked what humans remain better suited to do, Brady answered, “Problem solving, common sense, thinking with reason, understanding the big picture, understanding the context. Some physical tasks, as well.”
Humanoid robots remain in research mode
One area that appears less central for now is humanoid robotics. Agility’s Digit robot had a major appearance at the 2023 event, and Amazon has explored how bipedal robots might fit into fulfillment centers, including a pilot with Agility announced last year.
Since that pilot ended, however, Amazon has had little new to say publicly about humanoid robots. Brady confirmed that the partnership with Agility is still active, but he did not provide further details.
His explanation was cautious. “We’re still learning,” Brady said. “It’s slow and steady. ‘R&D’ is the best way I can capture it.”
The reason is not simply technical ambition. Brady said Amazon starts with the problem it wants to solve, rather than taking a piece of technology and looking for somewhere to force it into the workflow. In fulfillment centers, wheels work well on poured concrete floors, while stairs and uneven terrain become more relevant outside.
Machine learning is becoming more important to robot handling
Amazon’s robotics push also depends on better AI systems. One recent example is the company’s move involving Covariant, a UC Berkeley spinoff. In August, Amazon announced that it had hired Covariant’s founders, Pieter Abbeel, Peter Chen, and Rocky Duan, along with around 25% of its employees.
The goal is to expand the role of foundational models in industrial settings. That work is especially relevant for robotic manipulation, where machines must deal with an enormous range of products.
Amazon says Sparrow, one of its robotic arms, “can now handle over 200 million unique products of all different shapes, sizes, and weights.” Even so, warehouse automation still faces edge cases. Some will require human intervention, and others may be addressed by better-trained AI systems.
Covariant’s work with large datasets for product picking and placement fits directly into that challenge. Brady said of the deal, “We’re up and going and starting to work on some really meaty, very applied problems for machine learning.”
The overall picture is a robotics strategy built less around spectacle and more around integration. Amazon is scaling AMRs, robotic arms, storage systems, computer vision and machine learning inside facilities that must continue to ship packages. The future warehouse, in this version, is not a robot-only space. It is a dense, mixed system where machines handle more movement and manipulation while humans remain involved in maintenance, exceptions and decisions that require context.