Pathfinder

Exploring New Horizons

The Problem

What do you do when you are looking for a new job? A quick visit to Indeed, LinkedIn, or other professional platform usually is the start.

But that’s just the start.

Then comes the offer decoding, the profile tweaking, the application, etc. And that’s when you are the perfect fit for the job. But what about when you aren’t?

Enter Pathfinder.

Project Overview

The Career Pathfinder Engine is a data-driven decision system. Its goal is to bridge the gap between a job seeker’s current profile, market offer, statistical realities, and targeted educational opportunities.

How it works

Unlike naive job search engines, Pathfinder uses:

• Semantic job matches (no regex, keywords, etc)
• Semantic course matches (ditto)
• Semantic rating in relation to skills
• GenAI for CV tailoring
• Market metadata (where available)

The Market

Why do we serve market data?

Course recommendations based on semantic similarity alone don’t give the entire picture. We want to know how a course can potentially transform our life, in the context of what the market offers and the community’s opinion.

Segmentation

The StackOverflow’s survey covers a whole lot of technologies across several countries. We take care of extracting a picture of a localised reality as to avoid bunching together wildly different data groups.

For Germany:

For USA:

Which exposes the typical IT conundrum that sees fun techonology as the least in demand (by employers) and the most adopted technology among the least desirable (by developers).

The Pay Gap

How the StackOverflow’s survey distribute income across the the regions

How it looks like

Resources

GitHub repo at https://github.com/fourteatoo/pathfinder

GitHub Pages at https://fourteatoo.github.io/pathfinder

source: notebooks/slides.clj