NotebookLM-style App
<!--
Work info
-->
Role:
Full-Stack
Year:
2025
Local AI Knowledge Base
A self-hosted NotebookLM-style application built with Python and Next.js — turning any document collection into a queryable, LLM-powered knowledge base. Designed for privacy, portability, and local-first deployment.
Our Approach
Built around a RAG (Retrieval-Augmented Generation) architecture — documents are ingested, chunked, and embedded into a vector store. Queries retrieve semantically relevant chunks, which are passed to an LLM for grounded, context-aware responses. Fully Dockerized for cross-device portability with zero cloud dependency.
Key Features
Document ingestion & preprocessing pipeline
Embedding generation & vector search
LLM-based response generation (RAG)
Local-first, fully Dockerized deployment
Next.js frontend for document management & querying
Cross-device portability with no external API dependency
