“Training was all the rage for the last few years,” Atero founder Alon Yariv told TBPN in a recent interview. “People all of a sudden realize that they need to make money off of AI and inference is the only place that you actually make any money building a model.”
Inference, the process of serving a trained model so it can respond to user requests, is where the economics of AI are decided. It powers chatbot reply, every image generation, every automated email suggestion. The cost of delivering those outputs at scale is what determines whether AI companies can make money. Yariv’s team built Atero around that problem, focusing on memory as the bottleneck. “Models are huge. They’re about 1000 times larger than standard containerized applications. Just how much time it takes to put it into a GPU,” he explained. On top of that, user sessions can weigh gigabytes. Atero’s solution was a unified memory layer that moves assets across a GPU cluster quickly, allowing utilization rates to rise and latency to fall.
This week, Crusoe acquired Atero in a deal estimated at $150 million. The company is keeping Atero’s 25-person team together and turning its Tel Aviv office into Crusoe’s first presence in the Middle East. The acquisition comes as Crusoe positions itself as one of the most aggressive infrastructure providers in the AI sector. Founded in 2018, the company first became known for using stranded energy from oil and gas fields to power crypto mining operations. Over the last two years it has pivoted toward AI, raising more than $600 million to build large data centers and launching its own cloud platform designed for machine learning workloads.
“A lot of the technology that Alon and his team have built at Atero is incredibly complementary to our infrastructure software stack at Crusoe Cloud,” CEO Chase Lochmiller told TBPN. “From an efficiency standpoint, it’s a major investment. From a reliability standpoint, it’s a major investment.” He said the ability to manage memory more effectively can reduce inference costs dramatically. “If you simply optimize how you manage your KV cache, how the model gets managed, then you can actually drive down cost per token by half,” Lochmiller said. That measure, cost per token, is increasingly the benchmark that matters to customers. GPU hours are one thing, but the real question is how much it costs to generate a million words or process a million prompts.
The need for efficiency has only grown as models scale and as customers become more price sensitive. In May, Crusoe was announced as a partner in OpenAI’s Stargate project, a vast data center campus in Abilene, Texas. The project, expected to reach 1.2 gigawatts, has already secured more than $11 billion in funding. It will host as many as 50,000 Nvidia Blackwell GPUs per building. That buildout underscores the pace at which demand for compute is rising, but also the strain on economics. Without software to keep GPUs running close to full capacity, vast amounts of expensive hardware sit idle.
Yariv compared Atero’s approach to the kind of virtualization that transformed enterprise IT two decades ago. “The way I like to think about it is as the native virtualization for this new world. It’s virtualizing memory, which is the main asset. So I think about it as the equivalent of VMware for the wave of AI,” he said.
Lochmiller, who spent a decade in high frequency trading before founding Crusoe, drew his own analogy. “I do think there are elements of this that rhyme with what’s happening in high frequency trading,” he said. “As performance gains increase, utility goes up, and demand goes up with it. It has this reflexivity.”
The Atero team, which includes former OpenAI infrastructure engineer Ben Chess, now joins Crusoe’s broader engineering organization. Yariv described the combination as the fastest way to scale the technology they had been developing in stealth. “What we realized is that memory is the main bottleneck for this infrastructure, for inference,” he told TBPN. “That’s what we have solved with our technology. We have the unified memory layer. This allows you to utilize the GPU cluster to its fullest, shift memory assets fluidly and at the highest possible speed throughout the cluster.”
For Crusoe, the acquisition continues a pattern of building vertically. The company controls power sourcing through its energy operations, designs and constructs its own data centers, runs a cloud service, and now adds a specialized software layer to increase GPU efficiency. By keeping those pieces in-house, it can manage both performance and cost. Establishing a base in Tel Aviv also gives it access to Israel’s pool of high performance computing talent, a region that has become a hub for AI infrastructure startups.
Lochmiller summed up what the acquisition means for Crusoe, “Our goal here is to invest in the platform and drive massive performance gains that ultimately enable our customers to drive more value for themselves by using Crusoe Cloud.”
