Worked on backend and system-level improvements across the NVIDIA/KAI-Scheduler and langchain-ai/langgraph repositories, focusing on performance, resource management, and documentation clarity. Delivered a Go-based optimization to cluster information caching in NVIDIA/KAI-Scheduler, reducing memory allocations and processing time by refactoring object handling, which improved scalability for large Kubernetes clusters. Addressed GPU resource management by fixing MIG resource scheduling bugs, ensuring accurate allocation for MIG-enabled nodes. Enhanced onboarding and user experience in langgraph by correcting documentation grammar and clarifying workflow instructions using Markdown. Demonstrated strengths in backend development, Kubernetes scheduling, and system programming, with a focus on maintainability and efficient resource utilization.