🤖 Job seeker agent.ai for Public Services in New-Caledonia

#devchallenge #agentaichallenge #ai #machinelearning

This is a submission for the Agent.ai Challenge: Productivity-Pro Agent (See Details)

❔ What I Built

Currently, the way to find a job in Public Services in New-Caledonia is a classical form:

But when I come to look for a job this way:

"I'm a full stack developer but also passionate about data-engineering, AI & DEVOPS"

... things get complicated - time consuming 😵 - to find the best possible matches.

🎯 Why I built this agent

I built New-Caledonia Public Services (DRHFPNC) Job Seeker Agent to provide a ready to-use as a service agent that points anyone the best job opportunities according to its skills or aspirations so they find them & apply to them efficiently without having to crawl pdfs and run many "legacy searches".

🔭 Envision and perspectives

🗞️ Build a Job trends Blog Poster that makes weekly notes on specific skills and related job opportunities
📑 Custom job recommendation team : a Team that takes in input curriculum/skills from a specific set of sources (curriculum, online linkedIn activity,...) to make dedicated recommendations
🌐 Plug it into a team next to LinkedIn Profile Analyzer
📜 Create an agent that make is possible to put together all the assesments of a given user

Then it became public :

🍿 End User UX on mobile

Below a demo of the UX from a mobile device:

🔬 How it's built demo

💬 `Agent.ai` Experience

At first, I watched the video tutorial which took me a very few minutes, then I was ready to go to build the firs version of the agent.

🔬 How does it work ?

I have developped and scheduled a Kaggle Notebook : ⚙ Loader Avis de vacances de poste AVPs DRHFPNC that takes in input an OpenData dataset, performs transformations and as well pushed data into a Github Repo adriens/odata-avps-drhfpnc:

It pushed csv files and the markdown version from initial pdf ones.

For this agent I used the scraping feature of the raw csv file :