Team58_Penguin

Cross the mountain. One skill at a time.

Comment

Inspiration

The modern job market looks generous on the surface—thousands of listings, endless courses, countless “career guides.” Yet for students and early-career professionals, clarity is still painfully rare.

We noticed a recurring pattern: people weren’t failing because they lacked motivation or talent—they were failing because they didn’t know what to learn, in what order, and for which role. Existing platforms simply aggregate opportunities. None explain how to cross the gap between where you are and where a job expects you to be.

Penguin was inspired by the metaphor of a penguin crossing a mountain—not the fastest creature, not the strongest, but one that survives by taking deliberate, informed steps. We wanted to build that same kind of guide for careers.

What it does

Penguin is a personalized career navigation platform that turns fragmented career data into clear, actionable roadmaps.

It helps users:

Understand which roles actually fit their background

Learn exactly which skills to acquire—and in what sequence

Discover jobs linked directly to preparation roadmaps

Track applications and interviews in one place

For recruiters, Penguin:

Generates role-specific preparation roadmaps automatically

Improves candidate-role alignment

Reduces screening noise and application mismatch

In short: Penguin doesn’t just show you jobs—it shows you how to qualify for them.

How we built it

Penguin is built on a modern, modular, AI-first architecture designed for clarity, scalability, and fast iteration.

Frontend

React 18 + TypeScript + Vite

Tailwind CSS with shadcn/ui (Radix primitives)

Framer Motion for subtle, meaningful animations

TanStack React Query for data orchestration

React Router v6 for navigation

Backend

Flask (Python) for core APIs

FastAPI microservices for AI job ingestion and analysis

Firebase Firestore for user profiles and roadmap storage

CSV + JSON pipelines for flexible job data ingestion

AI & Data

OpenAI (GPT-4o-mini) and Anthropic Claude for:

Resume parsing

Job understanding

Skill dependency analysis

Roadmap generation

Job data normalized from:

eFinancialCareers

Glassdoor

JobStreet

MyCareersFuture

Indeed

Every job, skill, and roadmap is generated from centralized, structured data, not isolated heuristics.

Challenges we ran into

Fragmented data: Job listings across platforms used wildly inconsistent formats and terminology.

Skill dependency modeling: Determining which skills must come before others required careful AI prompting and validation.

UI complexity: Representing multi-stage roadmaps visually without overwhelming users was non-trivial.

Balancing AI autonomy and trust: We designed AI outputs to be transparent and interpretable—not black boxes.

Accomplishments we’re proud of

Built a fully integrated pipeline from resume → roadmap → job → application → interview

Designed role-specific learning paths instead of generic skill lists

Created a recruiter-side workflow that benefits both candidates and hiring teams

Unified jobs, learning, and application tracking into one coherent system

Delivered a polished, production-quality UI within a hackathon timeline

What we learned

Career clarity matters more than content volume

Good AI products are opinionated but explainable

Visual structure dramatically impacts user confidence

Data normalization is as important as model intelligence

The hardest problems aren’t technical—they’re about alignment and trust

What’s next for Penguin

Real-time skill readiness scoring

Resume-to-roadmap auto-updates

Recruiter analytics on candidate preparedness

Collaborative learning paths for cohorts

Deeper feedback loops from hiring outcomes

Penguin started as a way to cross the career mountain. Next, we want to make sure no one has to climb it alone.

Built With

Share this project:

Updates