SDE IV - GPU Engineer

Glance is an intelligent shopping agent, redefining the commerce experience. Powered by proprietary agentic intelligence and generative AI, Glance delivers a hyper-personalized consumer experience across mobile and TV — shaping the new era of shopping. Glance is operated by Glance InMobi Pte. Ltd., a non-consolidated subsidiary of global technology leader InMobi, and is backed by Mithril Capital, Google and Jio Platforms. To learn more, visit glance.com.

InMobi Group is a global technology company shaping the future of agentic commerce and advertising. Through its ecosystem of businesses — including InMobi Advertising and flagship consumer platform Glance — InMobi leverages data, machine learning, and generative AI to help brands reach audiences more precisely and consumers discover products more intuitively. Glance, which is pioneering new models of agentic commerce, is owned and operated by Glance InMobi Pte. Ltd., a non-consolidated subsidiary of InMobi Pte. Ltd.

InMobi Advertising, part of global technology company InMobi, is an agentic advertising platform helping brands and merchants achieve their business outcomes. Through its proprietary intelligence, AI-led solutions, and vast consumer reach — including flagship consumer platform Glance — InMobi Advertising delivers the omnichannel performance defining what's next in advertising and commerce. Glance is owned and operated by Glance InMobi Pte. Ltd., a non-consolidated subsidiary of InMobi Pte. Ltd. To learn more, visit advertising.inmobi.com.

As a GPU Systems Engineer, you’ll lead design and optimization efforts across our GPU inference stack.
You will architect the libraries and runtime systems that enable Stable Diffusion, multimodal transformers, and emerging video generation models to run efficiently at scale.

You’ll guide cross-functional teams, influence hardware selection, and set the technical vision for GPU optimization practices across the company.

• →Architect high-performance inference runtimes, kernel dispatchers, and memory planners for large diffusion and transformer workloads.
• →Lead investigations into cross-GPU performance bottlenecks, communication overheads, and scheduling inefficiencies.
• →Drive multi-GPU parallelism strategies — model, pipeline, and tensor parallelization.
• →Establish company-wide GPU optimization standards, tooling, and SLIs.
• →Collaborate with research to design scalable implementations of novel architectures.
• →Mentor engineers in profiling, tuning, and low-level optimization.
• →Partner with hardware vendors and infra teams to maximize cluster utilization.

• 5+ years in high-performance computing, GPU runtime systems, or ML infrastructure.
• Proven expertise in CUDA / Triton / C++, with deep understanding of GPU scheduling, occupancy, register usage, and tensor cores.
• Experience building and maintaining distributed inference or training systems.
• Ability to design abstractions balancing flexibility and performance.
• Strong knowledge of NCCL, NVLink, PCIe, and interconnects.
• Familiar with profiling automation and performance dashboards.
• Excellent technical leadership and mentoring capabilities.

• Background in compiler-aided optimization (TVM, XLA, MLIR, Triton).
• Experience tuning Stable Diffusion or transformer inference pipelines.
• Exposure to heterogeneous compute backends (AMD ROCm, TPU, ASICs).
• Experience working with hardware–software co-design initiatives.
• Open-source or research contributions in GPU optimization