Summary
Dheeraj Mekala is a research scientist specializing in machine learning and natural language processing with nine years of industry and academic experience, currently working on Agents & Reasoning at Meta after completing a PhD at UC San Diego. His work bridges principled, data-driven methods and practical scale—ranging from weakly supervised text classification and contextualized seed expansion in his doctoral research to building large-scale CV and ML pipelines that processed hundreds of millions of messages in production. He has a proven track record through multiple research internships at Meta, Microsoft, and Amazon, including contributions to next-generation Llama models and tool-usage capabilities in LLMs. Comfortable shipping systems and research, he combines deep modeling expertise with engineering rigor (auto-scaling Docker/Kubernetes frameworks, Kafka integrations, monitoring). Notably, he focuses on solutions that reduce human labeling effort by leveraging metadata and motif patterns, reflecting a consistent theme of amplifying model performance with light human supervision. Based in San Diego, he sits at the intersection of academic innovation and product-grade ML engineering.
9 years of coding experience
8 years of employment as a software developer
Higher Secondary Certificate (Class 12th), MPC, Higher Secondary Certificate (Class 12th), MPC at Sri Chaitanya College of Education
University of California, San Diego
Indian Institute of Technology Kanpur
Secondary School Certificate, Secondary School Certificate at SPR school of excellence
English, Hindi, Telugu