Priyanka Gujar

About

Priyanka Gujar is a Data Scientist and Machine Learning Researcher specializing in scalable AI systems and large-scale data analysis. She develops protein embedding workflows and applies transformer-based models to accelerate enzyme function and chemical property prediction, leveraging modular pipelines, distributed computing, and high-performance infrastructure. Her work involves running large-scale experiments on thousands of protein structures, comparing model representations, and iterating on pipelines to ensure results are reproducible and robust for downstream biological discovery.

In addition to life sciences, her research explores LLM-driven knowledge graph pipelines for structured knowledge extraction and model benchmarking across diverse datasets. Other projects span end-to-end machine learning pipelines for production systems, retrieval-augmented generation for personalized learning, and multi-agent frameworks for data quality monitoring and repair. Across these projects, tasks include preprocessing massive datasets, training and evaluating models at scale, and deploying systems that maintain performance under real-world conditions.

Committed to advancing the practical and reproducible application of AI, her work integrates rigorous methods, high-performance computing, and modern machine learning frameworks. By combining research in biological data with broader AI projects, the goal is to develop robust, adaptable computational tools and insights that can support discovery and decision-making in a wide range of data-intensive applications.

Research Interests

Transformer-based machine learning, scalable ML pipelines, LLM-driven knowledge graphs, model benchmarking, protein embedding workflows, enzyme function prediction, reproducible AI systems, retrieval-augmented generation (RAG).

Education

M.S. in Data Science, Northeastern University
B.E. in Electronics and Telecommunication Engineering, Savitribai Phule Pune University