How I Chose The Tech Stack For Job Helper

A developer shares the architectural decisions behind Job Helper, a personal tool built to streamline job searching using LLMs. The system is decomposed into microservices: a Python/Flask scraping service, a Celery/Redis-backed AI service that generates tailored CVs and cover letters via OpenAI or Gemini, a GraphQL gateway, and a Remix frontend. PostgreSQL handles relational data, Docker orchestrates and isolates services, and Google Docs serves as the document editor. Security is enforced via HTTP-only cookies, Docker private networking, and API keys for service-to-service communication. The author balances practical choices with deliberate exposure to new technologies like Remix and GraphQL.