wtf is personal-knowledgebase?

IntelliSense Platform is a desktop application that turns a pile of documents, such as PDFs, Word files, spreadsheets, and images, into an organized and searchable knowledge base you can ask questions against. It was built specifically with researchers and analysts working on unmanned systems like drones and uncrewed vehicles in mind, though the underlying tool could apply more broadly to anyone doing document heavy research. The project's stated design idea is that it builds and keeps up a set of structured knowledge entries once, rather than answers being figured out fresh from scratch every time someone asks a question, which is how many similar AI search tools work. It draws on a pattern called LLM Wiki, originally described by AI researcher Andrej Karpathy, and extends it for this specific research use. Core features include a question and answer system that uses a knowledge graph alongside retrieval to ground its answers in the source documents, complete with confidence scores and links back to where information came from. A deep research mode lets the tool reason across multiple documents in several steps while showing its progress. It automatically builds a knowledge graph connecting related pieces of information using four different signals, groups related knowledge into clusters, and can flag when two sources contradict each other. All data stays on the user's own machine with no cloud dependency, and the tool supports keeping several separate research projects isolated from one another. Under the hood it combines a Tauri based desktop shell written in Rust, a Vue 3 frontend, and a Spring Boot backend in Java, with a MySQL database and FAISS for vector based semantic search. It can connect to several outside AI providers including DeepSeek, OpenAI, and Anthropic to power its question answering and reasoning features.