Summary
Vinay Agarwal is an AI Solutions Engineer in Cupertino with 14 years of experience building production-grade ML and big data systems. He specializes in LLM fine-tuning, prompt engineering, and distributed multi-GPU training, having deployed model-driven copy generation and FHA-compliant text pipelines for Zillow/Trulia. His engineering breadth spans data pipelines, microservices, serverless apps, and cloud infrastructure (Spark, Flink, Java/Scala, Python, AWS, Terraform), enabling end-to-end ML lifecycle delivery from data ingestion to inference. Vinay has hands-on experience with cutting-edge tooling—PyTorch, Hugging Face, PEFT/LoRA, DeepSpeed, LangChain—and has driven efforts to replace human labeling with fine-tuned “smaller” models targeting 99%+ accuracy. He blends rigorous academic training (IIT Kanpur, UMD MS, Wharton MBA) with product-focused execution, and even earned a Kaggle bronze in Zillow’s Zestimate competition while contributing SEO and analytics tooling. Notably, he built a distributed multi-machine training environment in AWS and launched neighborhood description generation on Trulia in 2024.
14 years of coding experience
The University of Maryland, College Park
Bachelor of Technology (B.Tech.), Bachelor of Technology (B.Tech.) at IIT Kanpur
Master of Business Administration (MBA), Master of Business Administration (MBA) at Wharton School, University of Pennsylvania
Hindi