Resume

Role: Served as an ML Engineer and Backend Developer on Covalent, a unified compute framework to streamline AI and HPC workflows. Played a key role in enhancing the framework's functionality and scalability by developing backend solutions.

Interactive Workflow Visualization: Enhanced debugging efficiency by developing an interactive directed-acyclic graph (DAG) modeling feature for ML workflows. Reduced debugging time by 90%, improving team productivity and workflow traceability.

Dynamic GPU Resource Allocation: Designed APIs to enable on-demand GPU resource management, optimizing computational performance and minimizing idle costs in high-compute environments.

Streamlined ML Deployment: Built FastAPI-based APIs for one-click deployment of ML app endpoints via the Covalent Cloud Server, reducing deployment time from 1 hour to 5 minutes, significantly improving development agility.

AWS Integration for Pipeline Support: Worked with AWS services (EC2, S3, Lambda, RDS, CloudWatch) to support deployment pipelines, monitor performance, and resolve critical issues, ensuring system stability and seamless operations.

Cross-functional Collaboration and Communication: Acted as a key liaison between internal teams and clients, ensuring clear communication of critical issues and aligning technical solutions with business objectives to maximize impact.

Additionally contributed to the Covalent project by participating in Hacktoberfest, successfully resolving 3 issues with accepted pull requests, and receiving goodies as an appreciation.

Received the "Best Budding Star" award for consistently exceeding expectations on the Covalent project.