Version: 0.1.0 (Agentic Mode)
Goal: Build a fully agentic pipeline for paper discovery, retrieval, embedding, and reasoning with persistent memory.
Researcher is an intelligent research assistant designed to search, retrieve, and reason over scientific papers autonomously.
Version v0.1.0 introduces a full agentic architecture powered by LangGraph, mem0, and structured tool orchestration.
I'm Morteza, a computer science student and AI enthusiast. This project is part of my journey into LLMs, retrieval-augmented generation, and building autonomous research tools.
- π Search for academic papers (e.g., from ArXiv)
- π§© Embed, store, and retrieve papers using Chroma
- π§ Rerank and reason with LLMs
- ποΈ Maintain short- and long-term memory via mem0
- βοΈ Use CAG (Cache-Augmented Generation) to minimize redundant LLM calls