kamalchibrani-ai/FinAdvisor

FinSight is a full‑stack app for retail investors (India-first symbols like RELIANCE.NS) providing portfolio tracking, market analytics and an agentic chat assistant. Intended for individual investors and developers who want a reference implementation of LLM agents applied to finance. Key components: Streamlit UI, FastAPI backend with streaming NDJSON, Neon Postgres for durable storage, optional Redis session cache, and Pinecone for semantic long‑term memory. It uses LangGraph for a tool-calling agent (OpenAI chat + embeddings), yfinance for prices/fundamentals, bcrypt auth, and versioned SQL migrations. Primary features: email/password auth, CRUD portfolio holdings, live marks and P&L, dashboard metrics and charts, multi-thread persistent chat with agent tools (prices, fundamentals, comparisons, returns math, optional news), and deployable via Render. Requirements include Python 3.12, OpenAI and Pinecone keys, and a Postgres instance.

AI-related keywords found in this repo's description, topics, or README summary — grouped by category. Each badge links to the corresponding ranking detail page.