Associate Director - HPC Infrastructure Architect

Overview

Own the high-performance computing (HPC) infrastructure architecture that scientific workloads run on, spanning on-premises and cloud, and translate scientific demand into capacity models, workload placement principles and platform patterns. This is an individual contributor architecture role.

What you'll do

• Own the HPC reference architecture: cluster topologies, CPU and GPU node designs, high-performance and tiered storage, interconnect fabric, schedulers and the software stack that supports scientific workflows.
• Define the scientific workload placement model: which workloads run on dedicated clusters, which burst to cloud, and which belong on AI platforms, based on performance, data gravity, cost and compliance.
• Produce and maintain capacity plans for compute, accelerators, storage and network, including the power, rack and cooling requirements that data centre and colocation planning depends on.
• Establish data-locality and data-movement principles across instruments, laboratories, storage tiers, archive and cloud, including replication and recovery approaches for irreplaceable scientific data.
• Translate scientific requirements into sizing, benchmarking and procurement input, and validate feasibility before engineering build.
• Maintain architecture decision records for HPC platform choices, connecting each to scientific outcome, cost and lifecycle consequence.
• Define the interface with AI infrastructure so that shared accelerators, storage and data pipelines are used coherently rather than duplicated.
• Ensure HPC designs meet security, quality and regulatory expectations and are operable by Infrastructure Operations after handover.
• Engage scientific communities and research IT partners as the trusted architecture contact for high-performance computing.
• Coach engineers and analysts, and act as a role model for the Novartis Values and Behaviours.

What you'll need

• 8+ years of experience designing, building, or operating large-scale High-Performance Computing (HPC) environments.
• Proven expertise in HPC platform architecture, capacity planning, and multi-cluster infrastructure design.
• Strong experience supporting Linux-based compute clusters in enterprise or scientific computing environments.
• Hands-on knowledge of workload schedulers such as Slurm, LSF, or comparable HPC scheduling technologies.
• Solid understanding of CPU and GPU infrastructure design, performance optimization, and benchmarking methodologies.
• Experience with high-performance storage solutions, parallel file systems, and high-speed networking technologies, including InfiniBand and Ethernet fabrics.
• Knowledge of containerized, reproducible scientific workflows and hybrid or cloud-based HPC environments.
• Familiarity with large-scale data management practices, including data replication, archiving, and infrastructure planning (power, cooling, and rack design).
• Fluent English, written and spoken.

Nice to have

• Experience supporting research or scientific computing communities.
• Able to work fluently with scientists and translate research needs into infrastructure terms.
• Rigorous, evidence-based approach to sizing and benchmarking.
• Long-horizon planning discipline.
• Strong collaboration across engineering, facilities and research organizations.
• Additional languages.

Details

• Location: Hyderabad (Office).