Shreya Shankar — Machine Learning Engineer and AI Researcher at UC Berkeley
Shreya Shankar is a machine learning engineer and PhD candidate in computer science at University of California, Berkeley. Her work focuses on building reliable AI systems, with a particular emphasis on large language models (LLMs), data quality, and evaluation frameworks.
Research Focus: Reliable and Practical AI Systems
Shreya Shankar develops systems that help developers and organizations effectively use AI for data-centric workflows. Her research areas include:
- LLM evaluation and alignment with human preferences
- Data quality and data-centric AI systems
- Frameworks for building reliable machine learning pipelines
- Human-in-the-loop AI systems
Her work bridges the gap between theoretical research and real-world AI applications.
Influential Research and Publications
Shreya has published papers at top-tier conferences, including:
- SIGMOD (data management)
- VLDB (Very Large Data Bases)
- UIST (User Interface Software and Technology)
One of her most notable works is:
- “Who Validates the Validators?” — a paper exploring how to align LLM evaluation systems with human judgment, addressing a key challenge in modern AI development.
