Dhruvil Bhatt

About Dhruvil Bhatt

I am a highly motivated and results-driven computer science professional with a passion for leveraging technology to solve complex problems. With a Master\'s degree in Computer Science from the University of California, Irvine, and a Bachelor\'s degree in Information and Communication Technology from DA-IICT, India, I have honed my skills in various areas of computer science, including software development, data analysis, and machine learning.My journey in the tech industry has been enriched by diverse experiences, from interning at renowned companies like Synaptics Incorporated and HuddleUp to conducting research at esteemed institutions like the Indian Institute of Technology (IIT)- Bombay, and the Institute for Plasma Research (IPR). These experiences have equipped me with a robust skill set and a deep understanding of cutting-edge technologies.In my roles, I have demonstrated my ability to tackle challenges head-on and deliver innovative solutions. Whether it\'s engineering an Azure-based calling application, leading a team in curating the largest open-source dataset for corporate credit rating, or designing efficient algorithms for synthetic image generation, I thrive on pushing the boundaries of what\'s possible.I am proficient in a wide range of technologies, including AWS, Azure, Java, Python, JavaScript, React.js, Node.js, SQL, NoSQL, and more. My passion for learning and staying updated with the latest advancements drives me to continuously expand my knowledge and skills.Outside of work, I am an avid researcher and have contributed to publications in prestigious journals and conferences. My research interests span across areas such as computational finance, parallel computing, and computational plasma.As I continue to grow in my career, I am eager to collaborate with like-minded professionals and contribute to projects that make a meaningful impact. Whether it\'s developing innovative software solutions or conducting groundbreaking research, I am committed to pushing the boundaries of technology to create a better tomorrow.

Experience

1. Software Engineer (Machine Learning)

   Amazon Web Services (AWS)

   Oct 2024 — Present · Seattle, WA, US

   Designed a version-aware extension framework for vLLM that builds on its plugin architecture, enabling runtime-selective, version-guarded patches across 15+ LLM families on AWS Bedrock (Llama, DeepSeek, etc). This work was published on vLLM’s official engineering blog and featured on their LinkedIn page.• Architected a unified container build and validation pipeline for Bedrock and SageMaker LMI, usingEKS-based ephemeral environments to spin up and tear down endpoints for integration, chaos, and regression tests—reducing container count by 30% for AWS Bedrock and simplifying maintenance.• Extended the system with SageMaker-specific compatibility testing, automating Inference Component provisioning and validating LMI containers against runtime features (sticky routing, Multi-LoRA, orchestration) for full parity between EKS and SageMaker environments.• Built cross-region LLM benchmarking infrastructure that reduced latency bias by 30% and compute cost by 40% through ECS-based resource co-location and adaptive on-demand provisioning.

Education

• UC Irvine

  Master's degree, Computer Science

• Dhirubhai Ambani University

  Bachelor's degree, Computer Science (minor in Computational Science)

  2018 — 2022