Expert Review Challenge (Python)

This exercise is a 60-minute live, paired session in a small Python service. You will work in your own editor on your own machine and share your screen.

What to Expect

You will work through three parts:

1. Investigate and fix an existing behavior bug in the expert review queue API.
2. Extend the product with a small new review-flag feature.
3. Talk through what you cut, what you would want reviewed, and how you would harden the solution.

What We Are Assessing

We are interested in how you work, not just whether the app ends in a perfect state. We care most about:

• judgment under ambiguity
• code reading and debugging intuition
• technical quality and tradeoff awareness
• communication

Setup

Requires Python 3.12+.

python -m venv .venv
source .venv/bin/activate
pip install -e ".[dev]"

Run the server:

uvicorn expert_iq.main:app --reload --port 3001

Interactive docs: http://localhost:3001/docs

Run the tests:

pytest

No database, Docker, or external services are required.

Layout

• src/expert_iq/ — FastAPI service
  • main.py — app factory and ASGI entrypoint
  • routes.py — HTTP layer
  • service.py — business logic
  • repository.py — in-memory store
  • schemas.py — Pydantic request/response models
  • fixtures.py — seeded experts
• tests/ — visible test suite

Working Style

• Think out loud when you can.
• Ask clarifying questions when they materially affect your approach.
• It is fine to make reasonable assumptions and state them.
• It is fine to leave small polish items unfinished if you can explain the tradeoffs.