Supermemory Raises $2.6 Million Seed Round to Build AI’s First Long-Term Memory Infrastructure
Supermemory, an AI infrastructure startup building a universal memory layer for artificial intelligence applications, has raised $2.6 million in seed funding to accelerate the development and adoption of its long-term memory API designed to help AI systems retain and recall context across sessions. The funding round marks a pivotal moment for the San Francisco-area company, which was founded by 19-year-old entrepreneur Dhravya Shah and aims to address one of the most persistent technical challenges in the AI ecosystem: enabling models to remember and use information over time.
Supermemory’s core technology allows AI applications to ingest and process unstructured and multimodal data — including files, documents, emails, chats, PDFs, and streams from third-party apps — and build knowledge graphs that give models persistent, context-aware understanding. This capability can be applied to a wide range of use cases, from productivity and communication tools to more complex workflows in video editing, search, and robotics, all of which benefit from memory that goes beyond the short context windows typical of large language models.
The seed round was led by Susa Ventures, an early-stage venture capital firm with a focus on foundational technology and developer infrastructure. Joining Susa in the round were Browder Capital, a solo GP fund led by Joshua Browder of DoNotPay fame, and SF1.vc, a venture firm investing in forward-looking tech companies. In addition to institutional backers, the round attracted participation from influential individual investors in the AI and cloud computing space, including Cloudflare CTO Dane Knecht, Google AI chief Jeff Dean, DeepMind product manager Logan Kilpatrick, Sentry founder David Cramer, and executives from OpenAI, Meta, and Google, underscoring confidence from leaders with deep domain expertise.
Shah’s journey with Supermemory began as a side project called Any Context, a tool that allowed users to interact with Twitter bookmarks, which he built during a challenge to create new projects weekly. As the prototype evolved, it became clear that the underlying infrastructure for a scalable, persistent AI memory layer did not exist, prompting Shah to pivot the project toward a broader technology offering. His experience working on AI infrastructure and developer relations at Cloudflare also influenced the company’s technical direction and its focus on building reusable, high-performance memory components from scratch.
Supermemory’s product includes both a developer-oriented API and features such as a chatbot interface, presence in apps like Google Drive, OneDrive, and Notion, and a browser extension that enables users to add memories from websites. By converting unstructured inputs into organized, queryable memories, the platform aims to help applications deliver more relevant, personalised responses without requiring repeated reminders or manual context injection — effectively giving AI applications a form of “long-term memory.”
The funding comes as interest in AI memory technologies grows alongside broader investment in AI infrastructure. Supermemory’s early customer base includes AI tools such as the a16z-backed desktop assistant Cluely, video-editing AI Montra, AI search tool Scira, a multiproductivity tool Rube from Composio, and real estate startup Rets. The startup is also exploring partnerships with robotics companies seeking to retain visual memory from physical environments, demonstrating the versatility of its memory layer across both software and embodied AI use cases.
Investors have pointed to the strategic importance of memory for next-generation AI applications, noting that as models become more sophisticated, the ability to maintain continuity and context over time will be essential for user satisfaction and performance. Browder Capital’s Joshua Browder described his decision to invest as driven by Shah’s speed of execution and the product’s potential to become foundational infrastructure for AI developers.
Looking ahead, Supermemory plans to deploy its new capital to expand its engineering team, improve performance and scalability, and continue refining its multimodal data processing engine. With memory emerging as a competitive frontier in AI development, the company’s early funding and high-profile investor support position it to be a key player in the infrastructure layer that underpins future AI experiences.
