I have an existing Flask-based career-guidance site called MargDarshak and I’m ready to add five concrete improvements while keeping the codebase lean and easy to grade for my final-year project. The stack is already Python with Flask, so please stay within that ecosystem. Data lives in MongoDB, and I’d like any new collections or schemas documented clearly. Core additions I need you to build and wire into the current app: 1. Resume upload & parsing • Accept PDF, extract skills, education, and experience fields with a lightweight library (pdfminer, PyPDF2, or similar). • Store parsed fields in MongoDB alongside the raw file path. 2. Skill-gap analysis • Compare the parsed (or manually entered) skills against a chosen career profile to highlight gaps. • Return the gap list as JSON so it can be rendered on the front end. 3. Personalised career roadmap • Generate a step-by-step roadmap based on user data and identified gaps. • Persist the roadmap so progress can be updated later. 4. Progress tracking • Allow users to tick off roadmap items or mark skills as learned, then visualise completion percentage. 5. Feedback system • Simple five-star rating plus optional comment box for any recommendation or roadmap item. • Store and surface average ratings. Acceptance criteria • All new endpoints follow REST conventions and are covered by minimal unit tests. • Code is clean, commented, and separated into blueprints/modules so an academic reviewer sees clear structure. • A short README explains setup, new routes, and how to demo each feature. • Everything runs locally with `flask run`, requiring only MongoDB and the usual Python dependencies—no external AI services. If this scope is clear and you have prior Flask + MongoDB experience, let’s move forward—I’m on a tight academic timeline and would like steady progress updates.