I built a Prompt Node that converted brief style cues into detailed moodboard prompts and generated example outputs in n8n.

I built and delivered a complete Market Trend Discovery Agent, packaging data pipelines, AI processing, visuals, and a final report.

I built and assembled a Competitor Monitoring AI Agent that scraped product data from Wayfair, Amazon, and Walmart and produced a consolidated intelligence report.

I implemented planned enhancements to an AI Insights & Content Agent to make outputs more on-brand, complete, and production-ready.

I summarized how large language models work, explaining their architecture and training process.

I prepared and cleaned pharmaceutical documents with Python, using Pandas and image libraries to ready data and scans for AI processing.

I extracted and structured field-level data from multi-page SDF PDFs using PyMuPDF, producing a documented output with bounding boxes and extraction notes.

I compared Tesseract, PaddleOCR, and EasyOCR on a scanned pharmaceutical supplier PDF and produced a side-by-side evaluation report.

I built and optimized a RAG pipeline that indexed a short technical PDF, combined hybrid retrieval with reranking, and produced grounded answers to specific document questions.

I ran comparative RAG experiments and evaluated open-source LLMs on technical SDF documents, measuring retrieval precision, reranking effects, and model latency.

I built a page-level segmentation and classification pipeline to split multi-document PDFs and tag each page with doc_type and doc_id.

I built a Gradio-based document Q&A chatbot interface and connected it to a local RAG pipeline, including UI controls, streaming status, and a sources panel.