How Amazon cut development time of new Blue Jay robot

In fulfillment centers across the U.S., Amazon has been rolling out new technology that works with its employees to solve complex challenges, from delivery speed to sustainable AI. This week Amazon unveiled Blue Jay and Project Eluna, two developments it said build safer and more efficient workflows for its front-line employees.

“Our latest innovations are great examples of how we’re using AI and robotics to create an even better experience for our employees and customers,” says Tye Brady, chief technologist for Amazon Robotics. “The goal is to make technology the most practical, the most powerful tool it can be—so that work becomes safer, smarter, and more rewarding.”

Blue Jay combines 3 operations into 1 robot

Blue Jay is a robot that coordinates multiple arms to perform picking, stowing, and consolidating tasks simultaneously. This technology effectively collapses three assembly lines into one, Amazon said, creating greater efficiency in less space while supporting front-line employees.

Amazon said Blue Jay can pick, stow, and consolidate approximately 75% of all the various types of items it stores at its sites.

Amazon said Blue Jay’s development moved from concept to production in just over a year. It said the development process took three or more years for earlier robots like Robin, Cardinal, or Sparrow. Amazon said it was able to reduce the development time by reducing years of trial-and-error into months thanks to advancements in AI.

“Our engineers were able to iterate on dozens of prototypes for Blue Jay with the use of digital twins. These are an advanced form of simulation that now allow us to experiment virtually, using real physics to accelerate what we build,” Amazon said. “Combined with the AI, data, and learned experiences of our current robot fleet, we’re able to build systems like Blue Jay smarter and more quickly.”

Blue Jay is being tested in South Carolina. Amazon said it allows employees to shift from repetitive physical tasks like stowing items to higher-value work like quality control and problem-solving, making jobs less physically demanding while ensuring customers receive the right products faster.

Amazon earlier this year deployed its 1 millionth robot at a fulfillment center in Japan. It joined the retailer’s global network, which it said spans more than 300 facilities worldwide. Amazon also recently introduced its DeepFleet foundation model it said will improve the travel times of its robotic fleet by 10% and enable it to deliver packages to customers faster and at lower cost.

Project Eluna brings agentic AI to the warehouse

Project Eluna pulls in historical and real-time data across a building to anticipate bottlenecks and keep operations running smoothly. | Source: Amazon

Project Eluna is an agentic AI model that transforms how fulfillment centers operate. This digital tool processes real-time and historical data from across facilities, providing insights in natural language to help operations teams make better decisions.

Project Eluna enables operators to anticipate bottlenecks and chart efficient paths forward instead of scanning dozens of dashboards. Currently being deployed at a Tennessee fulfillment center for the holiday shopping season, Project Eluna will optimize sortation and eventually support preventive safety measures, helping plan ergonomic employee rotations and improve maintenance schedules.

The system allows operators to spend less time analyzing dashboards and more time coaching teams, creating safer work environments while enabling smarter, faster decision-making across Amazon’s global operations network.