itsa10but
An open-source résumé reviewer where four AI agents compare your résumé with a real job, back every score with a quote and point you to free resources.



Project Details
About the Project
Your résumé is a 10, but is it a 10 for this role?
itsa10but, a play on the "She's a 10 but" meme, is an open-source résumé reviewer that uses multi-agent AI to analyse your résumé against a real job and show you how to improve it. It's a one-page career workspace for students and people moving into tech and tech-adjacent roles.
Rationale
Many people know their résumé could be better but don't know where to start. The feedback they find is either generic ("add more numbers") or a score with no reason behind it. Tools that rewrite the whole résumé with one fixed model produce something that reads well but isn't really yours.
So itsa10but shows you where to look and leaves the writing to you.
How It Works
You add a job posting and your résumé, or one of the samples for eight role paths. Four agents then work in order:
- Role Scout — lists the skills the role asks for;
- Experience Finder — looks for proof of each skill in your résumé;
- Recruiter — checks ATS readiness and gives a quick review; and
- Action Planner — turns the biggest gaps into practical tips and free lessons.
A lightweight RAG step ranks résumé passages for each skill by word matching, with no embeddings or vector database. Every quote a model returns is checked against your text, and a quote that isn't found word for word scores 0. Scores describe proof on the page, never your chance of being hired.
Native Scan and API Boost
By default, a native scan runs the same four readers on rules and word matching in your browser, with no key needed. API boost runs the agents on the server with your own Anthropic, OpenRouter or Gemini key, and you can pick a model for every agent or for each one.
On 43 hand-written résumé and role pairs, the native scan finds the skills a résumé shows about 95% of the time and gives strong proof to real project work about 90% of the time. I treat these as a direction, not a benchmark.
Finding What to Learn Next
Every gap on the skills radar links to the Level up library of 116 free and open-source resources, with a small project to build and a résumé line to adapt. The workspace also shows the published values of 14 employers, Singapore hiring figures and AI headlines refreshed nightly through Firecrawl. Results download as a PDF report or a Markdown prompt for your own AI.
Résumés are read in your browser and nothing is stored. Emails, phone numbers and addresses are masked before anything leaves the page, and in API boost only that masked text goes to the provider behind your key.
Project Satisfaction
I built this from my belief that your résumé is personal and is worth more when you write it yourself. The workflow tells you specifically where to look, pairs you with the resources you may need and leaves the writing to you. I'm most proud of it because it is personal, genuine and cute.
Biggest Challenge
I'm a perfectionist, and I spent a long time designing the cute characters, layout and UI before the functions. Testing an app that depends on paid API keys, and making it real-time, was also hard. I overcame both through compound engineering: breaking big problems into smaller, solvable tasks and making careful changes so I didn't waste costly API calls.
Areas for Improvement
With more money and API keys, I would train my own LLM to represent the four agents, with sub-agents for MCP and API calls. My ultimate goal is real-time results and accuracy, and I hope to grow itsa10but into an official open-source tool that other reviewers can build on.
Build a project with us
Join AI Lodge, build with your track through the semester, and show your project at Hack Day. We will feature it here.



