Node.js LLM Trading MVP Build -- 2

Our team is assembling the first-release version of a trading platform focused on fast, reliable trade execution and management. The core stack is Node.js with Express, and an LLM layer will power intelligence features such as natural-language order entry, portfolio insights, or risk explanations—how you wire that up is part of the challenge and the fun.

Scope and goals
• Build a production-ready MVP in roughly four weeks.
• Develop REST/GraphQL endpoints in Node.js/Express that execute, amend, and cancel orders.
• Integrate an LLM (OpenAI, Anthropic, or similar) to interpret user intents and provide contextual feedback.
• Handle real-time market data and order routing for stocks and cryptocurrencies.
• Design for two user segments: individual traders and institutional desks, keeping roles/permissions clean from the outset.
• Keep the codebase test-driven, well-documented, and ready for future microservice break-out.

Acceptance criteria
1. A deployable server that can place demo trades to a sandbox broker or exchange for both stocks and crypto.
2. LLM interaction that translates at least three natural-language intents (e.g., “Buy 200 AAPL at market”) into valid API calls.
3. Basic dashboard or Swagger/Postman collection to showcase the endpoints.
4. Unit and integration tests covering critical trade flows; CI script passing on push.
5. Clear README with setup, environment variables, and next-step recommendations.

Assessment
https://github.com/david-wise66/backend-ai-assessment

We value your time, so please don’t spend too much time on the assessment. We’re mainly looking for your high-level thinking and approach, not a fully polished solution.

If you’re fluent in Node.js, comfortable around financial data, and excited about weaving LLM capabilities into trading workflows, I’m ready to dive into details and get this sprint underway.