GetThatJob

How "Get That Job" helps you out
Upload your resume and the job description of the aspiring job
5 Step process that generates you Insights
Showing you, how the ATS would view your resume
Changing your words into Impactful experiences

Inspiration

I come from India, where every graduate vies for a handful of roles in a vast talent pool. I’ve felt the sting of rejection, sending out dozens of resumes and never hearing back. I watched friends with strong credentials vanish into that hiring black hole, powerless to know why.

What if candidates had the same data and insights companies do? What if a friend, or a system could guide each bullet point, highlight missing keywords, and quantify achievements so every resume commanded attention?

A decade ago I shifted from Mainframe development into sales, product, and strategy. Building this vision pushed me far beyond my comfort zone. But today, thanks to GenAI and tools like Bolt.new, that challenge feels like a blessing: the very technology that once seemed arcane now powers our mission.

This isn’t just a technical feat, it’s deeply personal. Every job changes a life and uplifts a family. I set out to turn application anxiety into actionable confidence, offering job seekers a pathway from doubt to hope.

What it does

Dual Upload & Deep Analysis
Upload your resume and the job description. Instantly get a personalized Fit Score and a breakdown of missing keywords, skills, and role-specific gaps.
AI-Powered Bullet Improvements
Transform every experience line: weak verbs become strong action words, missing keywords weave in naturally, and achievements get quantified to pack maximum impact.
Personalized Action Plan
Receive a clear, prioritized list of exactly what to change—ranked by impact—so you know where to focus first for the biggest boost.
Polished PDF Report
All insights compile into a professional, multi-page PDF you can download or share—no data dumps, just a clean, executive-style consultation document.

How I built it

I’m a non-technical “vibe coder,” so I leaned into what I do best—strategy and prompts. First, I wrote a crystal-clear PRD outlining user personas, core flows, and success metrics before touching any code.

Then I handed the PRD and prompts to Bolt.new. In minutes, it spun up an end-to-end mock flow so I could focus on the real work: integrating our backend, tweaking the UI, wiring in the LLM, and squashing bugs.

I treated each AI prompt as its own mini-spec. I’d draft a prompt, test the output, refine it, enhance it with Gemini using discuss mode in Bolt, and repeat until I had the desired outcome.
For Deployments and integrations, I relied on Youtube and Bolt-Discuss mode to guide me.

With Bolt I felt empowered like Captain America with his shield, Thor with his hammer, even Thanos with the ring.

Challenges we ran into

LLM Hallucinations & Scope Creep
Our AI kept wandering—adding features I never asked for, fabricating details, even rewriting prompts. Taming it required enforcing tighter prompt chains and reminding the model of our PRD at every step.
Token Burn & Debug Logs
Bolt.new burned through tokens chasing bugs, so I sprinkled debug logs across the UI and backend. Then, in discuss mode, I’d trace the conversation to the exact file or function causing trouble. That became my fastest way to squash errors.
The Side Quest That Didn’t Ship
Mid-hackathon, I dove into a massive feature only to realize it wouldn’t make the deadline. It stung, but it taught me the value of focus. I’m already planning to pick it back up post-event with Bolt handling the heavy lift.

Each time I finally cracked an issue, I felt a rush of relief and empowerment, a reminder that even a self-styled solo “vibe coder” can build something real.

Accomplishments we’re proud of

A Truly Seamless User Journey
From the moment you hit “Upload” to the final action plan, the flow feels effortless. I managed to hide all the complexity—LLM calls, scoring logic, PDF generation—behind a clean UI that anyone can navigate without a manual.
The Gold-Standard Report
The PDF doesn’t read like raw data; it feels like a premium consultation deck. Admittedly, my first export still falls short of my own high standards, so post-hackathon I’ll double down to make it absolutely professional-grade.
That Moment of Surprise
Watching friends test the app and hear “Wait… you built this?” was priceless. Seeing genuine surprise—and a bit of awe—reminded me that a self-styled “vibe coder” can still pull off real magic.

What we learned

Prompt Engineering is an Art
AI is only as good as its prompts; iterative refinement directly improved our output quality.
Empathy-Driven Design
Putting ourselves in frustrated job seekers’ shoes guided every design and feature decision, ensuring genuine usefulness.
The Power of Narrative
Transforming complex data into a clear, actionable story proved the difference between a good tool and a great product.

What’s next for GetThatJob

This hackathon project is the foundation for a much larger vision. We aim to evolve from a single tool into an indispensable career co-pilot.

Feature Expansion
Integrate an Interview Prep tool and a Cover Letter Generator that aligns with both the resume and the job description.
Deeper Personalization
Train specialized models to deliver industry-specific advice and suggestions tailored to each user’s field.
Path to Market
Launch a Freemium model: a basic analysis for free to build community, with advanced AI rewrites and multi-report features behind a premium subscription.