Full-Stack Python Developer

Backend

• Strong Python (3.11+) with FastAPI or Django
• Async programming, background workers and task queues (Celery, RQ or similar)
• PostgreSQL and database design
• REST APIs, webhooks, idempotency, retries and signature verification
• Production web scraping: Playwright or Selenium, proxy rotation, rate limiting, session handling, and adapting to changing page markup
• OAuth flows and official third-party APIs
• One-time-code authentication by SMS and email, secrets management, and general application security

Front end

• Solid HTML5, CSS and JavaScript or TypeScript
• Embeddable widgets that work inside third-party pages: iframe and Shadow DOM isolation, CSP, and cross-origin messaging
• Mobile-first, accessible, performance-conscious UI
• Building dashboards and admin interfaces (React or Vue is fine)

AI

• Hands-on experience with the Claude API: prompt design, structured JSON output, image input, and cost and latency control
• Building evals and regression tests for prompts
• Designing around LLM uncertainty with confidence thresholds and human-review fallback

E-commerce platforms

• Shopify app development: Admin GraphQL API, checkout extensions, app billing, and mandatory privacy webhooks
• Working knowledge of WooCommerce (PHP and hooks), BigCommerce and Magento extension development
• Payment gateway concepts: authorize, capture, void and refund flows

Privacy and compliance

• GDPR and CCPA data handling
• Familiarity with marketplace and App Store review requirements

Experience and traits

• 4–7 years of full-stack experience, mid to senior level
• Takes ownership and is comfortable with ambiguity
• Pragmatic: ships a working version, then hardens it
• Designs defensively, assuming third-party sources will change and automated outputs will sometimes be wrong
• Communicates clearly and documents integrations
• Understands conversion-sensitive UX

You will build and extend features across the full stack of a web-based e-commerce product, working directly with the founder. You will own work from idea through production.

Screening and interview exercises (neutral)

Take-home (3–4 hours): Build a Python service that takes a JSON profile, calls Claude for a structured score with a confidence value, and routes low-confidence results to a review queue. Include tests. Design discussion: "A payment authorization is about to expire before a scheduled action completes. How do you detect it and handle it safely?" Front-end task: Build a small embeddable widget that survives a hostile host-page stylesheet, works on mobile, and calls an API without exposing credentials. Debug round: Diagnose a flaky scraper and a webhook that is firing twice.