A 19-year-old Nabs Backing From Google Execs For His Ai Memory Startup, Supermemory

Context windows of AI models, which bespeak nan expertise of a exemplary to “remember” information, person accrued complete time. However, researchers person suggested new ways to summation semipermanent representation of AI models, arsenic they often can’t clasp discourse complete respective sessions.

19-year-old laminitis Dhravya Shah is attempting to lick problems successful this area by building a representation solution, called Supermemory, for AI apps.

Originally from Mumbai successful India, Shah started building consumer-facing bots and apps a fewer years ago. He moreover sold his bot that formatted tweets into good-looking screenshots to nan societal media instrumentality Hypefury.

The founder, who was preparing for an entranceway exam to get into IIT(Indian Institute of Technology), earned bully money from this waste and decided to move to nan U.S to be Arizona State University instead.

After relocating, he challenged himself to build thing caller each week for 40 weeks. During 1 of those weeks, he built Supermemory (which was initially called Any Context) and put it connected GitHub. At that time, nan instrumentality allowed you to chat pinch your Twitter bookmarks.

The existent type of nan instrumentality extracts “memories” aliases insights from unstructured information and helps nan applications understand nan discourse better.

A representation representation extracted by Supermemory Image Credits: Supermemory

Shah secured an internship astatine Cloudflare successful 2024, wherever he worked connected AI and infrastructure. He later worked arsenic a developer relations lead astatine nan company. During this time, advisors, including Cloudflare CTO Dane Knecht, asked him to move Supermemory into a product.

This year, he decided to build Supermemory full-time.

Supermemory, now described arsenic a cosmopolitan representation API for AI apps, builds a knowledge chart based connected nan information it processes and personalizes nan discourse for nan users. For instance, it tin support querying crossed month-old entries for a penning aliases a journaling app, aliases searching for an email app. As nan solution allows for multimodal inputs, it could besides let a video editor to fetch applicable assets from a room for a peculiar prompt.

The startup tin ingest immoderate type of data, nan institution says, including files, docs, chats, projects, emails, PDFs, and app information streams. Its chatbot and notetaker characteristic lets users adhd memories successful text, adhd files aliases links, and link to apps for illustration Google Drive, OneDrive, aliases Notion. There’s besides a Chrome hold that lets you easy adhd notes from a website.

“Our halfway spot is to extract insights from immoderate benignant of unstructured information and springiness nan apps much discourse astir users. As we activity crossed multimodal data, our solution is suitable for each kinds of AI apps ranging from email clients to video editors,” Shah said.

Supermemory has secured seed backing of $2.6 cardinal led by Susa Ventures, Browder Capital, and SF1.vc. The information besides includes individual investors for illustration Cloudflare’s Knecht, Google AI main Jeff Dean, Deepmind merchandise head Logan Kilpatrick, Sentry laminitis David Cramer, and executives from OpenAI, Meta, and Google.

Shah said astatine 1 point, Y-Combinator besides approached him to subordinate 1 of its batches, but he already had investors connected board, truthful nan timing didn’t activity out.

Image Credits: Supermemory

Joshua Bowder, laminitis and CEO of “Robot lawyer” startup DoNotPay, who runs Bowder Capital arsenic a solo GP, was impressed by Shah’s tenacity.

“I connected pinch Dhravya complete X, and what struck maine was really quickly he moves and builds things, and that prompted maine to put successful him,” he said.

The institution has aggregate existing customers, including a16z-backed desktop adjunct Cluely, AI video editor Montra, AI hunt Scira, Composio’s multi-MCP instrumentality Rube, and Real property startup Rets. Plus, it’s moving pinch a robotics institution to clasp ocular memories captured by a robot.

While this is simply a slant towards consumers, nan app feels much for illustration a playground for developers to understand much astir nan instrumentality and perchance usage it successful their workflows aliases their ain apps.

Supermemory has important competitors successful nan representation space. Startups for illustration Felicis Ventures-backed Letta and Mem0 (where Shah worked for a short while) are building a representation furniture for agents. Supermemory’s ain backer, Susa Ventures, has invested successful Memories.ai on pinch Samsung, which tin pat into thousands of hours of footage to get insights. Shah says that while these startups mightiness service different industries and usage cases, Supermemory will guidelines retired because it offers little latency.

“More and much AI companies will request a representation layer. Supermemory’s solution provides precocious capacity while allowing you to aboveground applicable discourse quickly.” Bowder said.