The San Francisco startup is developing technology that allows AI systems to maintain context and relationships across large volumes of information.
Graphon Labs has raised $35 million in a Series A led by RRE Ventures, bringing its total funding to $43.3 million.
HP, Energy Impact Partners, Hitachi Ventures and Gaia VC also participated.
The startup was founded by Arbaaz Khan, Deepak Mishra and Clark Zhang.
Its central product is an AI memory infrastructure system the company describes as Persistent Intelligence.
Why AI Agents Forget
Large language models can process enormous quantities of information, but they do not inherently maintain perfect memory across changing enterprise systems and long-running tasks.
Traditional retrieval systems often break information into smaller pieces, potentially losing relationships and historical context.
Graphon uses a graph-based architecture designed to preserve connections between documents, conversations, images, videos and other data.
Instead of simply retrieving text relevant to a question, its technology attempts to reconstruct how information relates across sources and time.
Early Enterprise Adoption
The company reports customers and deployments involving the San Francisco 49ers, Hitachi and HP.
It is developing products for enterprises, software developers and devices requiring persistent memory.
Graphon also reports strong results on long-context memory benchmarks, although those findings should be interpreted within the conditions of each evaluation.
The Next Infrastructure Layer for AI
Reliable memory could become important as AI agents move from answering questions to performing work over days, weeks or months.
Graphon competes with database systems, retrieval infrastructure and memory capabilities being developed directly by major AI platforms.
Its challenge is proving that specialized memory infrastructure produces meaningful improvements in accuracy, efficiency and cost at enterprise scale.
Founders: Arbaaz Khan, Deepak Mishra and Clark Zhang. Investors: RRE Ventures, HP, Energy Impact Partners, Hitachi Ventures and Gaia VC.