The Real GPU Problem Isn't Cost. It's Scarcity.
At a time when GPU capacity is extremely limited, we explain how to get more impact from the GPUs you're already holding.
Pooja Malik
Distinguished Engineer
.webp%3Fv%3D2026-10-26T18%253A46%253A12.281Z&w=3840&q=75)
Bio
Pooja Malik is a Distinguished Engineer at Sedai, where she leads the work to make Kubernetes fully self-managing across AWS, Google Cloud, and Microsoft Azure. Since joining Sedai in 2021, she has shaped its core platform architecture and led its work on GPU optimization, autoscaling infrastructure, and observability pipelines. She is a recognized expert in GPU optimization for AI and ML workloads, with deep experience in DCGM telemetry, MIG partitioning, and Dynamic Resource Allocation. She also brings deep expertise in Kubernetes autoscaling (HPA, VPA, and KEDA) and large-scale observability with Prometheus, VictoriaMetrics, Grafana, and Loki.
Before Sedai, Pooja spent seven years at Salesforce as a Lead Software Engineer, where she helped make cloud infrastructure and database systems self-driving at scale. Earlier, she built high-scale systems at Walmart eCommerce, Visa, and E*Trade Financial.
Latest from Pooja