BizDaily

BizDaily tracks Indian startup funding news and turns it into a daily, readable brief. It pulls funding rounds from Inc42's public data API, uses an LLM to research and summarize each newly-funded company (problem, solution, traction, founders, risks, sources), and serves it all through a searchable table and a "notebook"-style daily digest.

Features

• Funding table — searchable, paginated feed of funding rounds, with a mobile-friendly card view.
• Daily brief — an LLM-generated Q&A style summary for every company funded on a given day, browsable by date.
• Company detail pages — deep dive on a single funding round plus its generated company profile.
• Bring your own LLM key — the backend only needs an OpenAI API key; no other accounts required to run it yourself.

Architecture

frontend/   React + TypeScript + MUI (3 routes: Home, Daily Brief, Funding Detail)
backend/    FastAPI, split into 3 mounted sub-apps (funding, brief, company details)
            + PostgreSQL for storage
            + OpenAI Responses API (with web search) for company research

See DEPLOYMENT_GUIDE.md for the full request flow and a step-by-step deploy walkthrough.

Prerequisites

• Python 3.11+
• Node.js 18+
• A PostgreSQL database (local, or a free one from Neon, Railway, Supabase, etc.)
• An OpenAI API key with access to the Responses API and web search

Local setup

1. Backend

cd backend
python -m venv venv
source venv/bin/activate      # Windows: venv\Scripts\activate
pip install -r requirements.txt

cp .env.example .env          # then fill in OPENAI_API_KEY and DATABASE_URL
python run.py                 # serves on http://localhost:4010

Tables are created automatically on first startup. Visit http://localhost:4010/docs for interactive API docs.

2. Frontend

cd frontend
npm install
cp .env.example .env          # defaults to http://localhost:4010, adjust if needed
npm start                     # serves on http://localhost:4011

3. Load some data

The funding table starts empty. Trigger a sync from Inc42 by hitting the refresh endpoint (or click Refresh Data in the UI once the frontend is running):

curl -X POST http://localhost:4010/api/funding/funding/refresh

This walks Inc42's paginated API and can take a while the first time. The Daily Brief page will generate company profiles on demand (via the LLM) the first time you view a given date.

Configuration reference

Variable	Where	Purpose
OPENAI_API_KEY	backend/.env	LLM company research (required)
DATABASE_URL	backend/.env	PostgreSQL connection string (required)
ALLOWED_ORIGINS	backend/.env	CORS allowlist for the frontend origin
REACT_APP_API_BASE_URL	frontend/.env	Where the frontend looks for the backend API

Deployment

See DEPLOYMENT_GUIDE.md for a free-tier deployment path (Render + Neon + Vercel).

Using a different LLM provider

backend/llm_service.py is the only file that talks to an LLM (currently OpenAI's Responses API with web search). To swap providers, replace get_async_client() and the client.responses.create(...) call with your provider's equivalent — the rest of the app only depends on get_company_details() returning a (company_name, sections_dict) tuple.

Contributing

Issues and PRs are welcome. Please open an issue before starting large changes so we can discuss approach first.

License

MIT