I’m building a new feature for BestKaam, our Indian job portal, and I need a production-ready resume parser that returns clean, well-structured JSON. The service must accept PDF, DOCX, and image resumes, detect text—even in scanned copies—and reliably pull out: • Personal details: name, email, phone, city • Professional timeline: job titles, companies, total experience • Skills, education, projects, certifications Accuracy on Indian résumé formats is critical, so feel free to combine regex, SpaCy, Transformers, NLTK, or any other NLP resources or any other tools that help you reach that goal. OCR with Tesseract (or a comparable engine) will be required for images and scans. Deliverables • Python codebase with clear modular design • RESTful endpoint (Flask or FastAPI) that accepts files and returns JSON • README covering setup, model training, and usage • Brief evaluation report showing accuracy metrics on a representative Indian résumé set When you reply, start with “I can build the BestKaam parser” and include links or descriptions of similar parsers you have delivered, your estimated timeline, expected accuracy, and best price. I’m ready to move quickly once I see a solid plan.