rAIsuMe

AI-powered exploration of my CV

Photo de profil

FAQ

Why rAIsuMe?
rAIsuMe is a technology showcase that demonstrates how AI can transform the exploration of structured data through a simple and intuitive interface.

Here applied to a CV, this approach is transferable to any type of content:

  • Technical documentation
  • Product catalogs
  • Corporate knowledge bases
  • Portfolios and complex files
The principle: coupling a clean conversational interface with AI to transform structured data (JSON, XML, databases...) into a natural user experience. No need to master complex interfaces or query languages anymore.

What works for exploring a CV today can automate access to your own data tomorrow, regardless of your field.
Which technologies are used?

This project was built using the following technologies and techniques:

  • Backend: Slim 4 (PHP framework), PSR-7 (HTTP messages), PHP-DI (dependency injection)
  • Frontend: HTMX for asynchronous interactions, Pico.css for responsive design
  • Data: CV in structured JSON Schema format, server-side storage
  • Document parsing: PDF Parser for extracting text from uploaded job postings (PDF)
  • LLM: OpenAI integrated via API with custom prompt system
  • Rendering: Markdown → HTML conversion of LLM responses
  • Internationalization: Custom translation service (FR/EN) with automatic language detection
  • Infrastructure: Azure Web App (production), Monolog logging
  • CI/CD: Automated deployment to Azure via GitHub Actions with tests and quality checks (PHPStan, PHPCS)
In which format is the CV stored?

The CV is stored on the server in JSON format. You can download the JSON source.

Can I paste a job posting?

Yes. You can either paste the posting in the prompt or attach a PDF or TXT using the “Choose file” button.

What instructions are sent to the LLM?

The instructions used to analyze the CV are defined in a system file. You can download these instructions as TXT.