Description
I need a developer for one small, self-contained job: building a single REST endpoint that wraps an LLM (OpenAI or Claude) for a basic customer-support Q&A. Deliverables: One POST endpoint in Node.js (Express) or Python (FastAPI) that accepts a JSON payload with the user question and returns the model's completion A well-structured system prompt for support context (I'll provide a page of grounding content) Integration with the Chat Completions / Messages API using sensible parameters (model, temperature, max_tokens) Error handling for empty input, rate limits (429), and API timeouts with proper status codes Environment-based API key handling (.env) and basic input validation A short README with setup and run steps Clean, commented code in a GitHub repo or zip To apply: Share one similar small task.