I compiled and submitted an Ideal Consumer List for a target segment during the externship module.

I analyzed survey data and produced quantitative and AI-driven qualitative deliverables, including spreadsheets and a written insights report.

I analyzed a mortgage document package to find missing items and proposed AI-assisted checks to automate verification.

I prepared and cleaned mortgage-related data and images using Python in Google Colab, producing notebooks and processed files for AI-ready document analysis.

I extracted and structured text and bounding boxes from multi-page mortgage PDFs using PyMuPDF, then documented issues and next steps for cleaning table and multiword fields.

I compared Tesseract, EasyOCR, and PaddleOCR on a scanned mortgage PDF and reported comparative findings and a recommended engine for automation.

I built a complete RAG pipeline for document retrieval and demonstrated it answering specific financial questions from a PDF.

I tested and compared three embedding models (MiniLM, E5, BGE) on mortgage policy queries and selected the best performer.

I built and tuned a page-level splitter and classifier that labeled pages and grouped them into documents for mortgage-related PDFs.

I built a Gradio chatbot interface that connected a PDF retrieval pipeline to a conversational UI and added features for file upload, summarization modes, and chat history export.

I identified a notable AR storytelling example and explained how it used AR to make news more immersive.

I started the Snap AR Lens Challenge by registering, installing Lens Studio, and preparing Snapchat for testing.

I produced a professional threat model for The Grand Marina’s HYDROLOGIC water management system.

I built Python scripts and Colab notebooks to practice core skills for IoT security data handling, including data structures, control flow, functions, JSON parsing, Pandas cleaning, and a sensor log generator.

I mapped an MQTT topic hierarchy and analyzed what data each topic would expose for a hotel's water system.

I prepared a security assessment report and experiments that validated TLS for Grand Marina Hotel's water sensor pipeline.

I documented a device provisioning and lifecycle policy for The Grand Marina's HYDROLOGIC system.

I ran experiments comparing timestamp checks, sequence counters, and HMAC signing, then wrote a recommendation for deployment.

I researched and explained how large language models (LLMs) work, focusing on tokenization, the Transformer architecture, and training vs fine-tuning.

I prepared and cleaned pharmaceutical document data using Python, applying Pandas, JSON handling, text standardization, and image preprocessing in Google Colab.

I built a Python-based extractor that pulled field-level data (dates, lot numbers, vendor, document type) from multi-page SDF PDFs and adapted regex to handle compact date formats.

I compared three OCR engines on a pharmaceutical PDF and reported which handled layout and text most effectively.

I built and tested a complete RAG pipeline for document retrieval, implementing embeddings, chunking, hybrid retrieval, and reranking for technical documents.

I tested three RAG configurations with an SDF document, compared retrieved chunks and reranker effects, and reported which setup produced the clearest grounding for answers.

I built a page-level splitter and metadata extractor that labeled each page in bundled PDFs with document IDs and types.

I built a Gradio-based document Q&A chatbot interface and connected it to a local RAG pipeline for interactive PDF question answering.