Prism — Query Intelligence Endpoint

What I built

• POST /queries -> accepts a natural language query, uses an LLM to extract structured intelligence (domain, geography, entity types, intent, filters), persists it in memory and returns the full record
• GET /queries/{id} -> retrieves a stored query record by ID
• Added a minimal HTML interface to submit queries and view extracted output

---

Note on LLM

I used Ollama cloud free (gemma3:27b model) via its OpenAI compatible instead of the Anthropic SDK as currently I don't have Anthropic API key.

---

Preview UI

---

How to run (for Ubuntu 24.04.4 LTS x86_64 )

• python -m env env -> create virtual environment
• source venv/bin/activate -> activate virtual environment
• pip install -r requirements.txt -> install all the requirements
• Copy .env.example to .env and fill in your Ollama credentials
• uvicorn main:app --reload --env-file .env -> to load env and run
• Open http://localhost:8000 or use the Swagger UI at http://localhost:8000/docs

---

One thing I would change with more time

Swap the in-memory dict for SQLite so query records persist across server restarts.

---