Staff Engineer, ML Profiling Tools

San Jose · SSI San Jose · USD 163,000-253,000 · 5+ years experience · Full-time · Onsite · Posted Oct 2, 2026

At a glance

The AGI Computing Lab at Samsung Semiconductor is seeking a Staff Engineer to develop interactive, large-scale web-based visual analytics tools for AI/ML workloads.
This role involves partnering with hardware architects and compiler engineers to translate complex performance metrics into intuitive developer workflows and diagnostic dashboards.

Develop and maintain highly interactive, large-scale and complex web-based visual analytics tools on AI/ML workloads.
Build interactive data visualization (flame graphs, call trees, timeline charts) capable of fluidly rendering massive, multi-dimensional infrastructure telemetry datasets.
Partner closely with compiler engineers and hardware architects to translate highly complex performance profiling metrics into clear, intuitive developer workflows.
Analyze and optimize the developer experience for machine learning performance profiling.
Design intuitive tooling and workflows to help engineers diagnose and eliminate inference bottlenecks.
Communicate effectively with stakeholders, including users, partners, and management, to ensure systems are delivered on time and within budget.
Complete other responsibilities as assigned.

Bachelor's with 10+ years, or Master's with 8+ years, or PhD's with 5+ years of industry experience.
Abundant experience designing and implementing rich, interactive data visualizations and user interfaces using modern web technologies.
Strong background or interest in creating developer tools, IDE extensions, or complex diagnostic dashboards.
Proven ability to translate massive, unstructured, or multidimensional telemetry/profiling data into intuitive, human-readable visual representations.
Strong track record of designing workflows specifically for technical users, focusing on minimizing cognitive load and streamlining root-cause diagnosis.
Hands-on experience with, or a strong curiosity about, systems profiling and tracing tools.
Baseline understanding of computing architectures (CPUs, GPUs, TPUs, interconnects/networking) and typical performance bottlenecks.
Excellent problem-solving skills and ability to think critically and creatively.

Working knowledge of modern machine learning frameworks (e.g., PyTorch, JAX, TensorFlow).
Working knowledge of execution paradigms (LLM inference, pipeline/tensor parallelism, kernel dispatch).