Adharvan Sai J

I’m an engineer at heart and a problem solver by nature. I specialize in HPC storage, large-scale AI/ML infrastructure, and distributed cloud orchestration, ensuring AI/ML workloads scale efficiently while maintaining high performance and cost optimization. As an SME in HPC storage, particularly Lustre, I focus on designing, optimizing, and troubleshooting high-performance distributed storage systems. I thrive on cross-domain challenges, whether orchestrating petabyte-scale data movement, optimizing compute-heavy AI workloads, or enhancing storage performance for distributed training, I build systems that work at scale, under real-world constraints. If it’s a tough problem, it’s worth solving. 💡 Why HPC & Large-Scale Infra? I design and optimize HPC, large-scale distributed cloud storage & compute infrastructure orchestration. My work ensures AI/ML training runs faster, scales efficiently, and remains cost-effective. When working with petabytes of data, high-throughput cloud storage, and compute workloads spanning thousands of CPUs and GPUs, performance is everything. My expertise spans: • HPC & AI/ML Infrastructure → Architecting scalable storage & compute for AI/ML training. • HPC Storage & Lustre → Optimizing high-performance Lustre-based storage systems. • Cloud & Distributed Systems → Designing cost-efficient, high-throughput AI/ML infra. • Storage & Compute Orchestration → Automating & scaling data pipelines & training workloads. • Performance Engineering → Debugging & optimizing latency, throughput & resource utilization. ⚡ How I Work—My Work Ethic & Mindset 💡 Problem Solver First, Specialist Second : I don’t just focus on one area—I solve problems wherever they arise. Whether it’s optimizing HPC storage, debugging AI/ML workloads, or resolving compute inefficiencies, I focus on getting to the root cause and delivering practical solutions. 🔍 Expanding Beyond Boundaries : The toughest challenges don’t fit neatly into one domain. I bridge gaps between storage, compute, AI, and distributed systems, constantly learning and adapting to solve problems that truly matter. 🔧 Hands-On, Not Just High-Level: I don’t just design systems—I build, debug, and optimize them. Whether it’s an OOM issue in AI training, a storage bottleneck in Lustre, or a misbehaving cloud workload, I dive deep to fix it. 📍 Constantly Learning, Constantly Building: The best engineers never stop learning. Whether it’s HPC advancements, AI/ML infra trends, or cloud performance optimizations, I’m always pushing boundaries and scaling systems beyond expectations.