From Confusion to Clarity: How Candogram and AI Fellows Are Streamlining Job Exploration

For students exploring potential careers, job titles can be more confusing than clarifying. That’s the problem Candogram, a visual job search platform, asked Break Through Tech AI Fellows to tackle in their recent Challenge Project.

Published
07/31/2025

At first glance, “Data Analyst,” “Business Intelligence Specialist,” and “Analytics Consultant” might sound like different jobs, but what if they’re more alike than they seem? For students exploring potential careers, job titles can be more confusing than clarifying. That’s the problem Candogram, a visual job search platform, asked Break Through Tech AI Fellows to tackle in their recent Challenge Project.

The project focused on transforming job titles from scattered labels into structured, student-friendly insights—making it easier to match interests with in-demand majors and careers.

Tackling the Language of Work

Employers often use different titles for similar roles, making job searches overwhelming—especially for students just beginning their career journeys. But underneath the branding, many job titles share overlapping skills, tools, and educational requirements.

Candogram’s challenge? Use AI to group similar job titles and create an interactive slider that allows users to fine-tune how broadly or narrowly they want to explore career options.

This isn’t just a technical exercise—it’s a tool for empowerment. By clustering job titles by similarity, the project helps students explore career paths based on real job postings, where companies often list preferred majors. The result: a clearer picture of how academic choices align with real-world demand.

AI Meets Career Guidance

The Fellows approached the problem as a research challenge. They experimented with word embedding models like Word2Vec and FastText to analyze the semantic similarities between job titles. Some even explored using large language models (LLMs) to build more nuanced comparisons. By feeding these models thousands of job titles, the team was able to group them into clusters that reflected underlying job functions—not just surface-level keywords.

Once the clusters were built, the Fellows developed a sliding scale interface, allowing users to adjust the similarity threshold. A tighter setting shows narrowly defined roles, while a broader setting reveals a wider variety of job titles with shared traits.

Reimagining Job Exploration

The Candogram project demonstrates how AI can make job exploration more intuitive and inclusive. For students deciding what to study or wondering where their major might lead, this tool becomes a map, translating the often opaque language of job titles into understandable, actionable information.

“Understanding the real-world demand for certain majors shouldn’t require guesswork,” said one Fellow. “We wanted to build something that helps students see what’s out there, even if they don’t yet know the right keywords.”

With Candogram and Break Through Tech, the job search becomes less about decoding jargon—and more about discovering careers.

Do you want to be part of our mission? Learn more about Break Through Tech’s AI Studio and the Challenge Projects that are helping more than 300 companies build an early tech talent pipeline.