A small, two-service project for researching subreddits and aggregating persona-style insights. The new structure separates a FastAPI backend from a Next.js frontend for clean deploys.

• backend: FastAPI service exposing /api endpoints
• web: Next.js 14 app (App Router) consuming the backend

Legacy notes, experiments, and older details have been moved to oldreadme.md.

• Node.js 20+
• Python 3.11+

• Backend (place in backend/.env):
  • OPENAI_API_KEY
  • REDDIT_CLIENT_ID
  • REDDIT_CLIENT_SECRET
• Frontend (place in web/.env):
  • NEXT_PUBLIC_API_BASE_URL (e.g. http://localhost:8000/api)

# Backend
cd backend
curl -LsSf https://astral.sh/uv/install.sh | sh
export PATH="$HOME/.local/bin:$PATH"
uv sync
uv run uvicorn main:app --reload --host 0.0.0.0 --port 8000

# Frontend
cd web
npm install
npm run dev

Open the app at http://localhost:3000. The API runs on http://localhost:8000.

Ensure you have backend/.env and web/.env populated as noted above.

• GET / — health check
• POST /api/research/subreddit — body: { subreddit_name: string, duration?: '1d' | '1week' | '1month', limit?: number }
• POST /api/research — legacy persona aggregator (kept for reference)

• Backend (Render)
  • Path: backend/
  • Build Command: curl -LsSf https://astral.sh/uv/install.sh | sh && export PATH=\"$HOME/.local/bin:$PATH\" && uv sync --no-dev
  • Start Command: export PATH=\"$HOME/.local/bin:$PATH\" && uv run uvicorn main:app --host 0.0.0.0 --port $PORT
  • Env Vars: OPENAI_API_KEY, REDDIT_CLIENT_ID, REDDIT_CLIENT_SECRET
  • Optional IaC: backend/render.yaml is included (monorepo).
• Frontend (Vercel)
  • Framework: Next.js 14 (auto-detected)
  • Env Var: NEXT_PUBLIC_API_BASE_URL → set to your Render backend public URL (e.g. https://your-render-service.onrender.com/api)
  • Build Command: default (next build)
  • Output: default (.next)

• The older notes and examples live in oldreadme.md to keep this README focused on the new structure and deploy path.

Sample curl:

 curl -v -X POST \
  -F "face=@/Users/sarthakagrawal/Desktop/agentData/sample.webp;type=image/webp" \
  -F "audio=@/Users/sarthakagrawal/Desktop/agentData/sample_small.mp3;type=audio/mpeg" \
  {{api_link}}/generateVideo \
-H "x-api-key: {{api_key}}" -o response.json

cat response.json | jq -r .video_b64 | base64 --decode > final.mp4
open final.mp4