Hilquias – Python, LLM, AWS

It's an end-to-end data extraction and transformation pipeline for processing large-scale telecom carrier invoices for a company that specializes in wireless expense management.

• Built a scalable pipeline to extract structured data from large PDF invoices using regex parsing and OpenAI function calling.
• Refactored and standardized carrier data pipelines using a Strategy Pattern for easier maintenance.
• Designed and implemented data marts to consolidate and normalize raw invoice data.
• Developed back-end endpoints to deliver validated, aggregated data for front-end visualizations.
• Improved cost efficiency and accuracy of invoice data extraction while ensuring data quality.

It's an intelligent women’s health assistant that supports users throughout their health journey. The chatbot integrates multiple capabilities, including answering complex medical questions, interpreting lab and imaging results, scheduling appointments, and more.

• Led a team of developers to design and implement a women’s health AI assistant.
• Created a RAG pipeline using curated health sources and validated medicine leaflets.
• Designed a user health database to centralize patient information for medical staff.
• Enabled chatbot capabilities, including Q&A, lab interpretation, and appointment scheduling.
• Implemented safety features for hazard detection and nurse escalation workflows.

It's a knowledge management platform that integrates RAG, role-based document access, and GenAI tools.

• Developed back-end features for a knowledge management platform with RAG and GenAI capabilities.
• Built a document creation pipeline combining user input, related context, and AI-generated summaries.
• Integrated GenAI tools for document editing, summarization, and text generation.
• Designed a robust PDF parsing pipeline to process tables for vector search and RAG accurately.
• Implemented role-based document management and secure feature delivery.

It's a multi-source data collection and analysis pipeline to track product information and market trends.

• Designed and implemented pipelines for scraping text, video, and image data from the web.
• Built a Named Entity Recognition pipeline to extract structured product features.
• Defined data models and applied OpenAI function calling for accurate structured extraction.
• Conducted market analysis on curated data to identify product trends.
• Developed an interactive Streamlit application with interactive plots and tables for client reporting.