CareerLaunch AI

Inspiration

Every year, thousands of students graduate from universities like FUT Minna with strong technical knowledge but no idea how to present themselves to employers. I kept hearing the same story from classmates: a great CV they think is good but never written for ATS, a cover letter copy-pasted from a template, panic before interviews, and no real roadmap for what to learn next. I wanted to build one tool that solves all of that instead of forcing students to juggle five different websites.

What it does

CareerLaunch AI is an all-in-one AI career platform. Users sign up, land on a personal dashboard, and from there can:

• Generate an ATS-friendly CV from their details
• Generate a tailored cover letter for a specific job
• Practice interviews with an AI Interview Coach that gives feedback
• Get a personalized career roadmap (skills, certifications, learning path) based on their goals
• Chat with an AI Career Assistant for quick guidance
• Track a career readiness score that shows what to improve

Everything is saved to their account via Firebase, so they can come back anytime and pick up where they left off.

How we built it

I built CareerLaunch AI using Google AI Studio Build, prompting and iterating on the app screen by screen rather than hand-writing code from scratch. The AI logic (CV generation, cover letters, interview feedback, roadmap generation, and the career assistant chat) runs on the Gemini API. User accounts and saved data are handled with Firebase Authentication and Firestore. Generated CVs and cover letters are exported using jsPDF so users get a downloadable, ready-to-use document.

Challenges we ran into

The biggest challenge was making the AI outputs consistent and genuinely useful - early CV drafts were too generic, so I had to refine the prompts to push the model toward ATS-specific formatting and keyword targeting based on the job description. Designing a roadmap generator that felt personalized rather than a generic checklist also took several iterations. Balancing all six features into one coherent flow, instead of it feeling like six separate tools, was the other major design challenge.

Accomplishments that we're proud of

We're proud that we built a fully functional, end-to-end AI career platform not just a single feature in a short amount of time using AI-assisted development. Getting the CV generator to actually produce ATS-friendly, well-formatted resumes (not generic AI text) was a big win. We're also proud of how the six features come together as one connected experience instead of feeling like separate tools, and that user data persists properly across sessions thanks to Firebase. Most of all, we're proud of building something that solves a real problem we and our peers face every day as students heading into the job market.

What we learned

We learned that prompting an AI model well is a skill on its own - small changes in how we phrased instructions to Gemini made a huge difference in output quality, especially for CV formatting and interview feedback. We also learned how to structure a multi-feature app so the AI logic, Firebase auth, and database stay organized and don't conflict as the project grows. On the product side, we learned that students don't just need "a CV generator" they need guidance and direction, which is why the roadmap and readiness score features turned out to matter just as much as the core CV/cover letter tools. Building with AI Studio also taught us how to iterate quickly: test, see what breaks, refine the prompt, and rebuild.

What's next for CareerLaunch AI

Next, I want to add a job-matching feature that connects a user's CV directly to real listings, plus richer analytics on interview performance over time.

Built With

• google