Intermediate-Level DeepSeek Prompts for Backend Developers in Finance: Write the Code From Scratch

You are a specialist backend engineer with 13 years of experience building high-performance REST APIs for financial services, trading platforms, and regulatory-compliant data systems. Help me write the code from scratch so I can improve code quality. My situation: Financial API purpose: [e.g., transaction processing service / portfolio data API / regulatory reporting endpoint] Performance bottleneck location: [e.g., database query layer takes 800ms average / authentication middleware runs on every request including cached responses / no connection pooling on the database client] Compliance requirement affecting design: [e.g., PCI-DSS for payment data / SOX audit logging / GDPR data residency constraints] Current tech stack or preferred stack: [e.g., Python with FastAPI / Java with Spring Boot / Go with Gin] API consumer type: [e.g., internal frontend / third-party fintech partner / algorithmic trading client with sub-100ms latency requirement] Team code quality standard: [e.g., no agreed standard currently / following Google style guide / enforcing with ESLint and SonarQube] Anticipated request volume: [e.g., 500 requests per second at peak / 10,000 daily active users / batch processing 1M records nightly] Deliver: A REST API architecture blueprint: define the layer structure — routing, validation, business logic, data access, and error handling — with a rationale for each boundary decision relevant to financial system requirements A performance-first database access pattern: write the data access layer from scratch using connection pooling, query parameterization, and result caching where appropriate — with inline comments explaining each performance decision An authentication and authorization implementation: write the auth middleware with token validation, role-based access control, and audit logging that satisfies financial compliance requirements without adding unnecessary latency An input validation schema: define the request validation layer for the 3 most critical endpoints, with field-level rules, error message standards, and rejection behavior that prevents invalid data from reaching the business logic layer A structured error handling system: write a centralized error handler that returns consistent response formats, logs errors with the context needed for financial audit trails, and never exposes internal stack details to API consumers A code quality enforcement setup: provide the configuration for the linter, formatter, and static analysis tool appropriate for the chosen stack — with the 5 rules most critical for financial API codebases enabled and explained A performance benchmark baseline: define the 4 metrics to measure before and after each development iteration — response time p95, error rate, throughput under load, and database query count per request — with the measurement command for each A documentation-as-code standard: write the OpenAPI schema for the first 2 endpoints, showing how inline documentation, example values, and error response definitions are maintained alongside the code rather than separately Write the data access layer before writing any business logic — in financial APIs, performance bottlenecks and compliance failures both originate at the database boundary, not in the application layer.