I mapped an Amazon fulfillment center workforce, detailing each role, responsibilities, tools, and reporting relationships.

I reviewed employee feedback examples and selected the most actionable Glassdoor reviews, explaining why each example demonstrated strong employee voice.

I cleaned noisy employee feedback text and produced a new cleaned_text column ready for analysis.

I built a keyword-tagged theme table and sentiment summary for employee feedback, highlighting burnout, leadership, and career growth across Glassdoor and YouTube.

I manually reviewed a merged pharmaceutical document package and recorded structural observations, date-format inconsistencies, and domain knowledge gaps relevant to AI processing.

I prepared and cleaned pharmaceutical document data using Python in Google Colab, then applied text and image preprocessing techniques to improve AI readiness.

I extracted structured data and key dates from pharmaceutical SDF PDFs using PyMuPDF and regex, then documented the results in a Colab notebook.

I compared Tesseract, PaddleOCR, and EasyOCR on a scanned pharmaceutical Certificate of Quality and reported which engine best preserved layout and key fields.

I built a complete RAG pipeline for document retrieval and produced a notebook and report showing retrieval results.

I ran three RAG configurations, logged retrieval differences, and identified the configuration that produced the most focused answers.

I built page-level segmentation and metadata tagging for bundled PDF files, producing structured document IDs and types for downstream routing.

I built a themed Gradio chatbot UI and connected it to a RAG pipeline for answering questions about an uploaded PDF.