Atombeam has raised $18.3 million through a Regulation A+ offering hosted on StartEngine, bringing the AI infrastructure company’s total funding to more than $55 million.
The company plans to use the capital to accelerate commercialization of Neurpac, its data compaction and encryption software, while advancing its Persistent Cognitive Machine, or PCM, through independent benchmarking and customer pilots. Atombeam said PCM completed its first full build in September.
Atombeam said its intellectual property portfolio includes 678 patent filings, with 308 issued or allowed and 370 pending. Neurpac is designed to reduce the volume of machine-generated data that must move across networks while allowing the original data to be restored at the receiving end.
According to the company, Neurpac can typically remove 50% to 90% of the data that needs to move in the use cases described in its announcement. Atombeam said the software can compact small packets that are difficult for conventional compression systems and increase effective bandwidth on infrastructure customers already own. The company cited deployments and validation across satellite, defense, and commercial communications.
Atombeam is separately developing PCM as an AI architecture with persistent memory and continuous learning capabilities. The company is targeting resource-constrained environments in which conventional large language models can be limited by computing, memory, power, or connectivity requirements.
Atombeam’s broader strategy combines AI-powered data infrastructure with a new model architecture designed for continuous learning. The company sees both technologies as responses to the growing network, data-center, and computing demands generated by connected devices and AI workloads.
KEY QUOTES:
“Closing this round marks an inflection point for Atombeam. We now have the capital, the team, and the technology to accelerate Neurpac adoption while bringing PCM through independent benchmarking and into customer pilots. Neurpac is in production with customers today, measured on their own traffic. PCM is something genuinely new: an AI that keeps learning after it ships and becomes an expert in the organization it serves with no data center behind it. In November we will put it beside conventional language models in independent benchmarks.”
“With Neurpac we are solving the challenges Large Language models exacerbate – wired, cellular and satellite networks and data centers pushed to the brink by heavy AI workloads and the barrage of data from connected machines. Simultaneously, with PCM we are creating a form of AI that addresses the very weaknesses of LLMs that lead to unprecedented network congestion and data centers that use as much power as small cities. In these ways we are using AI to address both the challenges of AI adoption today while creating the innovations required to realize AI’s full potential in the future.”
Charles Yeomans, Co-Founder, Chairman And CEO Of Atombeam
