Time Paradox

Macoostudy Career Agent: don’t just claim the skill. Prove you can defend it. Every resume makes technical claims, and interviewers test them. Most candidates only find out which claims they can’t defend in the real interview, when it’s too late. Macoostudy Career Agent is an AI technical skill auditor that finds out first. You upload your resume and the job description. Muse Spark extracts every concrete technical claim as an exact span of the resume, with its line, skills and competency. “Full stack” isn’t one skill, so each claim is tested against the competency it actually describes: backend, DevOps, ML, databases. The claims most relevant to the job come first. An adaptive technical interview then probes one rubric criterion at a time: implementation, design rationale, trade-offs, failure handling, measurement. Follow-ups build on what you actually said. Our Claim-to-Evidence Engine grounds every verdict. Evidence must be a verbatim quote from your answer, the model only labels criteria, and the backend computes every point. Same answers, same score. Each claim ends as Supported, Partially supported or Insufficient evidence. Gaps are split into knowledge gaps and communication gaps, because “I don’t know this” and “I can’t explain this” need different fixes. You get a resume heatmap that colours each claim by verdict and links to the answers behind it, a report with readiness and a competency breakdown, and a practice task for every gap. Instead of one-to-one mock interviews, it’s AI used the right way: you answer, it shows exactly where you’re going wrong, and you grow. It’s for candidates, not a screening tool. Built during the hackathon: the entire Career Agent: backend engine, 7 database tables, every /agent page, the new homepage and 37 tests. The earlier Macoostudy resume-roast tool predates the event and lives separately at /roast.