Project 1: How AI Reads and Understands Mortgage Documents

I explored the principles of Machine Learning, Deep Learning, and NLP to understand how AI reads mortgage documents. Additionally, I applied Computer Vision and OCR techniques, analyzing a real mortgage 'blob' file to evaluate AI model performance on structured and unstructured data.

Project 3: Data Extraction from Documents Using Python

I converted digital and scanned PDFs into machine-readable formats using Python. By employing tools like PyMuPDF and pdfplumber, I extracted key information from multi-page mortgage files, applying heuristics such as regex patterns to identify essential fields.

Project 5: Introduction to Retrieval-Augmented Generation (RAG)

I designed a Retrieval-Augmented Generation (RAG) pipeline using LlamaIndex to enhance information retrieval from large document sets. This project improved my understanding of how AI can efficiently access relevant data in extensive text collections.

Project 7: Blob Processing, Classification, and Routing

I developed a system to process and classify unstructured mortgage file blobs. By using layout-based clues and rule-based methods, I separated documents and built a routing function to apply appropriate extraction logic for each identified type.

Project 9: Final Integration, Testing, and Evaluation

It’s time to bring everything together. In this final project, you’ll integrate all components into a complete, end-to-end document intelligence system. You’ll run your pipeline on a full mortgage blob, from preprocessing and OCR to extraction, RAG retrieval, and user interaction. Then, you’ll rigorously test the system’s performance, evaluating accuracy, speed, and reliability across different document types. Finally, you’ll package your work into a clear, professional demo—including a Colab repo, documentation, and walkthrough video—that showcases your system’s real-world potential. By the end, you’ll have a fully functional AI-powered document automation platform ready to present to peers, mentors, or even hiring managers.

Project 2: Learn to Work with Python for AI-Powered Document Processing

I utilized Python to process and clean mortgage documents, focusing on data preparation techniques. I worked in Google Colab, gaining proficiency in Python basics and enhancing OCR accuracy through image processing methods to ensure data readiness for AI automation.

Project 4: Advanced OCR Comparison and Layout-Aware Extraction

I compared three OCR engines—Tesseract, PaddleOCR, and EasyOCR—by extracting text from complex mortgage documents. I learned to clean text and utilized layout-aware tools to maintain formatting, ultimately recommending the best OCR tool for automation based on performance.

Project 6: Advanced RAG and Open-Source Experiments

I advanced my RAG pipeline by implementing improved chunking techniques and metadata-based filtering. I evaluated the performance of open-source models, comparing them with proprietary options to optimize the pipeline for real-world mortgage document retrieval.

Project 8: Gradio Chatbot with RAG Integration

I created an interactive chatbot using Gradio that integrated my RAG pipeline. This project allowed users to query documents via a web interface, enhancing the usability of my retrieval engine and making it easier to switch between different AI models.