Amazon remains the largest manufacturer and operator of robotics worldwide. At its annual Delivering the Future event, held this year in San Francisco, the e-commerce giant unveiled its latest robotics system, Blue Jay, alongside a new agentic AI model, Project Eluna. “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,” said Tye Brady, chief technologist for Amazon Robotics, at the event. “The
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,” he added.
Designing for people
Blue Jay collapses three robotics stations into one workspace to pick, stow and consolidate approximately 75 per cent of the items it stores on its sites.
“Visually, Blue Jay operates like a juggler who never drops a ball – only here, the ‘balls’ are tens of thousands of items moving at high speed. It’s also like a conductor leading an orchestra, with every motion in harmony,” Amazon said in a statement.
It is designed to shift employees from repetitive physical tasks, such as stowing items, to higher-value work, such as quality control and problem-solving, while ensuring customers receive the right products faster.
The name of the latest next-generation robot follows the naming tradition of birds, like Sparrow, Cardinal, and Robin.
Like all of Amazon’s robotics, Blue Jay was built in Massachusetts and is currently being tested at a facility in South Carolina.
Amazon puts people at the centre of its robotics universe, referring to fulfilment centre employees as its customers through its Amazon Robotics arm.
“The next leap in AI is reasoning with friction and physics in the real world,” said Aaron Parness, director of applied science for Amazon Robotics at the event.
“The systems that win are the ones that learn from contact, the ones that make human work better because of their presence across everything we’re building,” he continued.
Designing at speed
According to Amazon, Blue Jay’s development moved from concept to production in just over a year. Nearly one-third the time of prior robotic projects including Sparrow, Cardinal and Robin.
Amazon shared that its engineers iterated on dozens of Blue Jay prototypes through the use of digital twins – an advanced form of simulation that enables virtual experimentation using real physics to accelerate development.
The company added that, when combined with AI, data and insights from its existing robot fleet, this technology allows it to design systems like Blue Jay with greater intelligence and efficiency.
“We had an amazing physics-based simulator so we could understand how to grasp things, and we could actually train the foundation model that’s running it through virtual means right through simulation alone,” Brady told Inside Retail.
“That was a really big deal for us, because it allowed us to really ascertain if the system then validated the overall architecture, a valid architecture,” he continued.
“Then once we build it, we try to minimise the gap, what we call ‘sim to real’ – what happens in the simulation does not necessarily happen in the real world.”
From Brady’s perspective, the Amazon Robotics team is only getting better at closing the simulation-to-real-life gap, enabling faster development timelines on projects like Blue Jay.
Blue Jay is slated to become a core technology for Amazon’s Same-Day Delivery sites, potentially shortening package delivery times for customers.
Further reading: Inside Amazon’s plan to hand its drivers smart glasses