Create a Pain Point Chart

I utilized AI tools such as Perplexity and ChatGPT to identify customer frustrations, creating a Pain Point Chart that highlighted key insights. This analysis informed the foundation for a comprehensive journey map focused on genuine consumer needs.

Analyze Consumer Insights

I analyzed consumer insights derived from survey data, conducting both quantitative and qualitative analyses. Using Google Sheets, I created data visualizations that presented findings clearly, culminating in AI-driven insights that deepened my understanding of consumer behavior.

Synthesize Consumer Insights

I focused on creating impactful presentations, learning slide structure and design principles. I utilized AI to plan and refine my presentation, incorporated charts with Gamma, and developed a data dashboard using Claude, ultimately presenting my findings in a video format.

Map the Customer Journey

I distinguished between consumers and customers, developing an 'Ideal Consumer List' tailored for targeted segments. I applied the AIDA framework to create a Customer Journey Map, effectively mapping crucial touchpoints between the business and its consumers.

Conduct Sentiment Analysis

I performed exploratory data analysis (EDA) and sentiment analysis using Python and Google Colab. I gathered and cleaned customer review data with Pandas and NumPy, visualizing insights that revealed consumer sentiment trends.

Project 1: How AI Reads and Understands Mortgage Documents

I studied how AI understands mortgage documents through machine learning, deep learning, and natural language processing (NLP). I analyzed the mortgage industry's challenges and worked with a real mortgage 'blob' file to see how AI processes both structured and unstructured data.

Project 3: Data Extraction from Documents Using Python

I extracted text from both digital and scanned mortgage PDFs using Python libraries like PyMuPDF and pdfplumber. I employed field heuristics, including regex patterns and layout clues, to identify key information such as borrower names and loan amounts.

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

I built a Retrieval-Augmented Generation (RAG) pipeline using LlamaIndex to enable AI to retrieve relevant data from extensive document sets. This project enhanced my understanding of integrating retrieval mechanisms into AI models.

Project 7: Blob Processing, Classification, and Routing

I designed a system to process unstructured mortgage blobs by classifying and routing them to appropriate extraction methods. I applied layout-based clues and rule-based techniques to identify different document types within the blobs.

Project 9: Final Integration, Testing, and Evaluation

I integrated all components into a complete document intelligence system, running the entire pipeline on a full mortgage blob. I rigorously tested the system's performance and created a professional demo, including documentation and a walkthrough video.

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

I utilized Python for data processing and cleaning of mortgage documents, preparing them for AI automation. I gained hands-on experience with Google Colab and developed techniques to standardize and enhance the accuracy of data for optical character recognition (OCR).

Project 4: Advanced OCR Comparison and Layout-Aware Extraction

I compared three OCR engines—Tesseract, PaddleOCR, and EasyOCR—to determine their effectiveness on complex mortgage documents. I learned to clean and extract text while preserving formatting, ultimately recommending the best tool for real-world document automation.

Project 6: Advanced RAG and Open-Source Experiments

I improved my RAG pipeline by implementing advanced chunking techniques for better document segmentation. I also utilized metadata-based filtering to enhance the relevance of retrieved answers and evaluated the performance of various open-source models.

Project 8: Gradio Chatbot with RAG Integration

I developed an interactive chatbot using Gradio that enabled users to query my RAG pipeline through a web interface. This project involved modularizing my code and integrating various LLMs for a seamless user experience.

Project 1: Getting Started with AI Agents - Moodboard Generator

I developed my first AI agent, a Moodboard Generator, which transformed prompts into visualizations. This project laid the groundwork for visualizing trends from abstract concepts, demonstrating how efficiently insights can be generated for application at Wayfair.

Project 3: Building the Competitor Monitoring AI Agent

I built a Competitor Monitoring AI Agent that collected and analyzed competitor listings and customer feedback. This agent generated actionable insights into competitor strategies and market positioning, enabling Wayfair to make informed merchandising and marketing decisions.

Step 1: Understanding the Dashboard Builder Agent

I focused on understanding the purpose and functionality of the Dashboard Builder Agent. Building on previous projects, I conceptualized how to integrate existing outputs into a single, decision-ready dashboard, gaining clarity on the core concepts necessary for the upcoming build.

Project 2: Building the Consumer Trend Discovery AI Agent

I created a Consumer Trend Discovery AI Agent that analyzed real-world data from sources like Amazon and design blogs. This agent effectively identified emerging product trends, allowing Wayfair to respond to shifts in consumer preferences more swiftly than traditional research methods.

Project 4: Designing the AI Insights & Content Agent

I designed a customized AI Insights & Content Agent that transformed market trends and competitor data into marketing content. By understanding Wayfair's brand voice, I enhanced the agent's output by improving tone and expanding content types, resulting in a tool that the marketing team could