Comment on page

Intro: The FirstBatch Stack

Overview & The FirstBatch Stack

FirstBatch stack is the set of technologies that are built around personalization and AI, creating the necessary tools that make next-level user experiences possible while also democratizing the rapid developments in the AI landscape, especially LLms.

User Embeddings

User embeddings are vector representations of users that store their behaviors, interests, preferences, etc. They are essential for enabling personalized and customized experiences across websites & applications, leveraging the unique properties of vector embeddings that allow efficient search and similarity calculation methods. When a user signs up, their personas & interests are calculated based on the snapshots taken from their social accounts. User embeddings are stored on the blockchain without a user id, wallet address, or email. Instead, users prove ID ownership in Zero-Knowledge to update their data, making it permissionless. Learn More


HollowDB is a ZK-enabled open-source decentralized key-value database on the Arweave network, developed by FirstBatch. With its high scalability and performance, HollowDB is perfect for serving as a decentralized LLM cache for AI-powered applications. The goal of HollowDB is to provide a database for applications with actual production-level stability and feasibility for projects of any scale, while also being privacy-preserving with Zero-Knowledge Proofs. Learn More

HollowDB Vector

HollowDB Vector stands as an open-source, decentralized vector database designed for bolstering the development of Large Language Model (LLM) applications with persistent memory, while also enhancing user privacy and ownership. Learn More