AI model context windows, which indicate a model’s ability to remember information, have grown over time. However, researchers are exploring new methods to enhance the long-term memory of AI models, as they frequently struggle to maintain context across multiple sessions. A 19-year-old founder named Dhravya Shah is addressing this challenge by developing a memory solution called Supermemory for AI applications.
Originally from Mumbai, India, Shah began creating consumer-facing bots and apps several years ago. He even sold a bot that formatted tweets into attractive screenshots to the social media tool Hypefury. The founder, who was preparing for an entrance exam for the Indian Institute of Technology, earned a significant amount from this sale and decided to move to the United States to attend Arizona State University instead.
After relocating, he challenged himself to build something new each week for forty weeks. During one of those weeks, he built Supermemory, initially named Any Context, and shared it on GitHub. At that time, the tool enabled users to chat with their Twitter bookmarks. The current version of the tool extracts memories or insights from unstructured data and helps applications better understand context.
Shah secured an internship at Cloudflare in 2024, where he worked on AI and infrastructure. He later served as a developer relations lead at the company. During this period, advisors, including Cloudflare CTO Dane Knecht, encouraged him to transform Supermemory into a full product. This year, he decided to work on Supermemory full-time.
Now described as a universal memory API for AI apps, Supermemory builds a knowledge graph from the data it processes and personalizes the context for users. For example, it can support querying across month-old entries for a writing or journaling app, or searching for an email application. As the solution allows for multimodal inputs, it could also enable a video editor to fetch relevant assets from a library based on a specific prompt.
The startup states it can ingest any type of data, including files, documents, chats, projects, emails, PDFs, and app data streams. Its chatbot and notetaker feature lets users add memories in text, add files or links, and connect to apps like Google Drive, OneDrive, or Notion. A Chrome extension is also available for easily adding notes from a website.
The core strength of Supermemory is to extract insights from any kind of unstructured data and give applications more context about users. Because it works across multimodal data, the solution is suitable for all kinds of AI apps, from email clients to video editors.
Supermemory has secured seed funding of 2.6 million dollars led by Susa Ventures, Browder Capital, and SF1.vc. The round also includes individual investors like Cloudflare’s Knecht, Google AI chief Jeff Dean, Deepmind product manager Logan Kilpatrick, Sentry founder David Cramer, and executives from OpenAI, Meta, and Google.
Shah mentioned that Y-Combinator approached him to join one of its batches, but he already had investors committed, so the timing did not work out.
Joshua Bowder, founder and CEO of the DoNotPay startup and who runs Bowder Capital as a solo GP, was impressed by Shah’s tenacity. He connected with Dhravya over X and was struck by how quickly he moves and builds things, which prompted him to invest.
The company has multiple existing customers, including the a16z-backed desktop assistant Cluely, the AI video editor Montra, the AI search tool Scira, Composio’s multi-MCP tool Rube, and the real estate startup Rets. It is also working with a robotics company to retain visual memories captured by a robot. While there is a slant towards consumers, the app feels more like a playground for developers to understand the tool and potentially use it in their workflows or their own applications.
Supermemory faces substantial competitors in the memory space. Startups like Felicis Ventures-backed Letta and Mem0, where Shah worked briefly, are building a memory layer for agents. Supermemory’s own backer, Susa Ventures, has invested in Memories.ai along with Samsung, which can analyze thousands of hours of footage for insights. Shah says that while these startups may serve different industries and use cases, Supermemory will stand out by offering lower latency.
As more AI companies require a memory layer, Supermemory’s solution provides high performance while allowing users to surface relevant context quickly.