How GreyOrange GreyMatter makes warehouse orchestration intelligent

It was a busy second half of 2025 for Saurabh Gupta.

In mid-July, the former Apple and Amazon executive took on a new role in the intelligent hardware and autonomous systems world, as he was named CTO of GreyOrange, the warehouse orchestration and inventory software provider.

Now, Gupta and the GreyOrange team have their sights set on 2026 for more deployments and advancements.

Historical data gives GreyMatter a head start

The scalable orchestration platform GreyMatter from GreyOrange gives customers a plan before ever being deployed in the warehouse.

After years of data collection and analysis, GreyMatter harnesses its computations and advanced technological capabilities to optimize processes right away.

“When GreyMatter is installed, it can start running and give you fairly close to your desired performance without any data,” Gupta said. “But then part of what it does in its early days is actually collect the data which will help improve the decision-making over time… The notion is not that you need to spend months or years to get to performance. You get performance on day one, but then the system is going to learn over time and continually adapt and optimize based on what is happening in the warehouse.”

The AI-powered platform, which GreyOrange said generates 1 million real-time optimizations per minute - across 100+ sites and 10,000+ robots, does so across various warehouse operations. GreyMatter is deployed in 3PL, logistics, fulfillment, e-commerce and other settings.

“Even though there would be aspects of GreyMatter which are fairly similar, there would always be domain-specific information that the client will care about,” Gupta said. “One of our goals with GreyMatter is, how do we combine the two? How are we able to get that learning on how palletization should happen - which is very specific to a retail distribution center - but combine that with navigation, which could have been learned generically across any installation, to then give the customer those combined insights.”

Certified Ranger Network adds robot flexibility

As a hardware-agnostic platform, GreyMatter provides prospective customers a degree of flexibility when it comes to the robots that it orchestrates.

From the GreyOrange perspective, Gupta said that the Certified Ranger Network (CRN) expands via an ongoing dialogue between the company, its customers and hardware providers.

“I would say there's a pretty healthy mix of all three,” he said. “Traditionally, the way it started was as you're working with customers, they told us, ‘Hey, we are working with or we are exploring this particular vendor, and we like their robots. Does it work with GreyMatter?’ Or, ‘Hey, can you talk to them and figure out what it takes to work with GreyMatter?’”

The ongoing relationship between GreyOrange and its customers helped drive more partnerships to the CRN, which supports the following platforms from six different robotics vendors:

• Tote-to-person automation
• Case-picking automation
• Intralogistics automation
• Rack-to-person automation
• Assisted picking cobots
• Sortation and smart conveying robots

“Before we had proven that we can do this, there was skepticism in the market,” Gupta said. “For a robot vendor, it's like, ‘Why do I need GreyMatter? Why can't I just work directly with the customer?’ That initial push from customers really helped. When one of these big players said, ‘Hey, we really like this particular robot. We really like GreyMatter. Can you figure out how to work together, and can we get the best of both?’ That was the seed for a lot of our partnerships.”

DeepNav partnership looks to reduce AMR deployments

One month after Gupta joined GreyOrange, the company announced a major collaboration with Google Cloud. The partnership will see the development of GreyMatter DeepNav, which the companies describe as an AI-powered offering for dynamically managing and optimizing autonomous robotic operations at scale.

As part of the announcement, GreyOrange and Google Cloud predicted that with GreyMatter DeepNav, AMR deployment times will drop by up to 80%.

How? Navigation.

“One thing I can share is that it will probably be the first deployed system in warehouses, which is actually using a transformer-based GPT model,” Gupta said. “We are effectively taking what people have done with text, everything that you see with Gemini and ChatGPT, and we have taken that and extended that to navigation.”

While elements of path-planning and movement optimization are critical for robotic navigation in warehouses, the predictive capabilities are what Gupta said will drive (pun intended) this platform.

“Just like with text in ChatGPT and Gemini and all of these agents, they are able to predict the next sentence or the answer to your question,” he said. “What we have done is similar with navigation. Given the state of different robots in the warehouse, our GPT system is able to predict what is the next set of actions or the next set of directions that the robot should move to get to the optimal output. There's still some work we need to do because it's a brand-new concept in terms of how navigation works and where we are currently.”

The initial news said that the platform will be available in early 2026. While Gupta was mum on the specific details, he hinted that exciting news is coming soon.

“We have proven it out in simulations that it can actually get to much higher performance compared to these traditional mechanisms,” he said. “There is still some more work we need to do in the real world to get this system to deal with all of the intricacies of what you see in real life and there's still some tuning we need to do to get it to be as robust as what ChatGPT is. We’re not too far off. Expect to see some more sort of tidbits and announcements for us over the next few months. It's a very exciting project for all of us.”