Technical Requirements & Stack Constraints

Overview

This skill defines the mandatory technical stack and constraints that must be followed when implementing the backend challenge.

It exists to prevent accidental deviation from the expected technologies and architectural style during development.

When to Use

Use this skill when:

• •Choosing languages, frameworks, or libraries
• •Designing application architecture
• •Writing Docker or infrastructure files
• •Making decisions about persistence, APIs, or services
• •Setting up project structure or tooling

Mandatory Stack

The implementation MUST use:

• •Language: TypeScript
• •Runtime: Node.js
• •API: REST (HTTP + JSON)
• •Database: PostgreSQL
• •Containerization: Docker
• •Local orchestration: Docker Compose

If an alternative is considered, stop and document why before proceeding.

Architectural Guidelines

• •The system MUST be backend-only (no frontend)
• •The application SHOULD follow a layered or hexagonal architecture
• •Domain logic SHOULD be isolated from infrastructure concerns
• •Business logic MUST NOT depend directly on framework-specific APIs

Persistence Rules

• •PostgreSQL MUST be the source of truth
• •Database schema SHOULD be explicit and normalized
• •Migrations MUST be used for schema evolution
• •Avoid storing core business logic inside the database

API Guidelines

• •APIs MUST be RESTful and JSON-based
• •Endpoints SHOULD be versioned if public-facing
• •Input validation MUST be explicit
• •Error responses SHOULD be consistent and structured

Containerization Rules

• •The application MUST be runnable via Docker Compose
• •Local development MUST NOT require external services
• •Environment variables MUST be used for configuration
• •No secrets should be hardcoded

Testing Expectations

• •Core business logic SHOULD be unit tested
• •Authorization combinatorics SHOULD have explicit test coverage
• •Tests SHOULD be runnable locally via npm/yarn scripts

Explicitly Forbidden

• •Using a different language or runtime
• •Introducing frontend frameworks
• •Using managed cloud services directly (AWS, GCP, etc.)
• •Tight coupling between domain logic and infrastructure
• •Skipping Docker setup

Example

User: “Should we use MongoDB for flexibility?” Assistant: No. PostgreSQL is mandatory per technical requirements.

User: “Can we use Python for faster prototyping?” Assistant: No. TypeScript + Node.js are required.