Skip to content
All work

Case study — AI financial analyst

JengaVest

An AI financial analyst platform with live market data.

Live demoSource code

How it works

  1. 01

    Ask

    Next.js

    A question about a stock, the market or a report.

  2. 02

    Retrieve

    ChromaDB + BM25

    Hybrid search over PDFs: vectors plus keywords.

  3. 03

    Enrich

    Finnhub

    Live stock prices and news add current context.

  4. 04

    Answer

    Claude API

    Claude reasons over the context and responds.

OutputGrounded answersLive market dataPDF analysis

The problem

Market data, news and company reports usually live in different places. JengaVest brings them together behind one conversational analyst.

What I built

  • Designed and built an end-to-end conversational analysis platform on the Claude API, combined with live stock and news data (Finnhub) for real-time market intelligence.
  • Architected a hybrid RAG pipeline (ChromaDB vector search + BM25 keyword retrieval) for PDF document analysis.
  • Served by a Python/FastAPI backend and a Next.js frontend.

Want something like this for your team?

Tell me about the problem. I'll tell you how I'd build it.

Email meAsk my AI