Hi, my name is

Pratik Antoni Patekar.

I build intelligent data systems.

I am a Research Scientist & Data Engineer specializing in scalable data infrastructure, RAG systems, and enterprise Generative AI applications. Currently building at UNC Chapel Hill.

About Me

I architect scalable data infrastructure and end-to-end AI applications, bridging the gap between complex data pipelines and advanced research systems.

Currently, I serve as a Research Scientist at the University of North Carolina at Chapel Hill's School of Data Science & Society (SDSS). I engineer enterprise data workflows and Generative AI systems, translating complex faculty research requirements into production-ready infrastructure that ensures data integrity and accelerates cross-disciplinary innovation.

Beyond the screen, you will likely find me speedcubing on my GAN 3x3 (current solve time ~30s), weightlifting, or hiking the trails around North Carolina.

Core Technical Stack

Data Engineering & Cloud

PySpark Microsoft Fabric Azure Databricks ETL/ELT SQL

AI & Machine Learning

RAG & GraphRAG Azure OpenAI LLMs Vector Embeddings

Languages & DBs

Python Neo4j PostgreSQL MongoDB

Featured Projects

GoFAR Platform

In Preview

Lead Architect

Architected a vector-based semantic search and clustering web application. Processed and indexed over 14,000 academic profiles using OpenAI embeddings, successfully facilitating 20+ cross-disciplinary research partnerships.

Azure OpenAI Vector Search Python

Multi-Agent HR Bot

In Preview

AI Systems Engineer

Designed a plan-and-execute multi-agent architecture utilizing domain-specific expert agents to handle complex university policies. Streamlined HR workflows by providing synthesized answers grounded in institutional data.

LangGraph Copilot Studio Python

Visual QA Model

Machine Learning Engineer

Engineered a CNN and LSTM deep learning model to answer counting questions based on images. Achieved 88.9% accuracy on the MS COCO and VQA v2 datasets, demonstrating CV and NLP integration.

TensorFlow Keras CNN/LSTM

Music SQL Backend

Database Engineer

Architected a robust relational database for a music rental business using Oracle SQL, leveraging advanced data modeling techniques to improve query efficiency by 50%.

Oracle SQL Data Modeling RDBMS

AI Policy Voyager

In Preview

Researcher and system developer

Created an interactive platform combining Vector RAG and Knowledge Graph retrieval to analyze legislation. Authored multi-hop Cypher queries to construct searchable graphs for complex legal data analysis.

Neo4j GraphRAG Cypher

Where I've Worked

Research Scientist

Oct 2025 - Present

UNC Chapel Hill, SDSS

  • Directed the "SDSS Computational Data Science & AI Program," providing technical consultation and managing cloud compute resources to accelerate HPC adoption for 30+ faculty members.
  • Architected and deployed scalable GenAI prototypes leveraging Azure OpenAI and RAG methodologies.
  • Co-authored technical sections for AI/ML grant proposals, securing 3 UNC AI Accelerator Program awards totaling $100,000 in research compute funding.

Data Engineer

Mar 2024 - Sep 2025

UNC Chapel Hill, SDSS

  • Architected a centralized data ecosystem using Microsoft Fabric, designing complex ETL pipelines to unify 5+ data sources and reducing data access latency by 40%.
  • Built secure, scalable cloud data processing pipelines that served as foundational infrastructure for the university's first 3 high-impact generative AI deployments.

Data Engineer

Nov 2023 - Mar 2024

NEXTGEN Innovation Labs

  • Executed flawless migration of 25,000 customer records across networks using advanced ETL operations, achieving 95% transition success and preventing 60+ hours of downtime.
  • Engineered optimized data workflows in SQL and Python, cutting processing time by 40%.

Education

Master of Science

GPA 4.0

Computer Science

The University of Texas at Arlington

Aug 2021 – May 2023

Bachelor of Engineering

GPA 3.74

Electronics & Telecomm

The University of Mumbai

Aug 2015 – May 2018