Celonis LLM Backend

This is the Celonis LLM Backend. REST API written in Python using the Flask framework.

What is the Celonis Challenge

In Celonis Challenge we had to recreate the application process. The goal was to create a tool that would help recruiters to find the best candidates for a job position. The tool is able to automatically generate questions for the candidates based on the job description and the CV of the candidate. Additionally, it is able to automatically generate a CV for the candidate based on the answers and the original CV. Finally, the tool is able to calculate a seniority level and similarity score to the job description.

Files structure

The main functionalities can be found in cloud_function/api.py and cloud_function/functions.py, and they include the following endpoints:

• /get_questions: get auto-generated questions for given job description and CV of candidate to present to the candidate in the frontend
• /get_metrics: get metrics such as main_information (name, email, Nationality), detailed information from original CV (education, work experience, skills), and information from the answers (Experience, Skills). Additionally, it calculates a seniority level based on the candidate's answers/CV.
• /get_autogeneratedcv_info: automatically generate a unified CV based on candidate's CV and the metrics above, with the addition of a similarity score based on the similarity between the skill set of the candidate (CV + Answers) and the one of the Job Description.