How brunotanabe/curriculum-vitae-query-assistant works
Project type and purpose
• This is a backend-only, containerized AI microservice (“curriculum-vitae-query-assistant”) that exposes a Swagger-enabled REST/HTTP API to:
• Ingest resume files (PDF, JPEG, PNG)
• Run OCR (via EasyOCR or PyPDF)
• Summarize/extract content using an LLM (via HuggingFace)
• Answer job-fit queries
• Log usage metrics (no document storage)
• It follows a hexagonal (ports-and-adapters) and domain-driven structure with two main apps: core and curricula_vitae.
File structure
• Top-level:
• setup/: ASGI/WGSI entrypoints and Django/FastAPI settings
• apps/core/: core abstractions (boundaries, controls, entities, utils)
• apps/curricula_vitae/: domain-specific code (API views, services, DI container, entities, serializers, migrations)
• Dockerfile & docker-compose.yaml: containerization
• manage.py, pyproject.toml/requirements.txt: Python project and dependencies
• The “boundaries” directories define ports (interfaces to OCR/LLM/DB), “controls” implement use-case logic and standard response handling, “entities” hold enums and data models.
README for additional insights
• Confirms OCR + LLM services, Dockerization, Swagger API, and logging (likely to a relational DB via Django migrations).
Detailed guidance
• Outer adapters (controllers & external-service implementations)
• Inner core (use-case controls & entities)