AI Career: Design and Technical Documentation of a Browser-Based, Human-in-the-Loop AI Tool for Resume Analysis, Job Matching, and Interview Preparation

Job seekers must simultaneously maintain an accurate resume, search several job boards, tailor application materials, track applications, and prepare for interviews. Large language models (LLMs) can assist with each task, but they also introduce risks that matter in hiring contexts, including fabricated qualifications, opaque scoring, and loss of user control over what is sent to employers. This paper describes the design and technical implementation of AI Career, a free browser-based career-coaching application published by AI Verse. The application combines transparent deterministic components—a rubric-based applicant-tracking-system (ATS) readability check, a dictionary-based skill extractor, and a weighted job-match score—with generative components for resume parsing, job-fit evaluation, resume tailoring, cover-letter drafting, interview-question generation, answer scoring, and conversational coaching. A multi-step career agent searches job sources (Remotive, Arbeitnow, and optionally Adzuna), ranks postings, evaluates the strongest matches, prepares tailored materials, and places every proposed change to the application tracker in an approval queue. Application autofill is implemented as a user-approved bookmarklet that fills form fields but never submits an application. All user data are stored locally in the browser, and AI requests are sent only to the provider the user selects: a free hosted service, an AIV key, Groq, any OpenAI-compatible API, a local model, or an in-browser model. The paper documents the architecture, scoring algorithms, features, and a step-by-step tutorial workflow, and discusses limitations and safeguards. No formal evaluation of match validity, ATS outcomes, interview performance, user satisfaction, or employment outcomes has been conducted; the benefits described are proposed rather than demonstrated.