ML Accelerator Performance Engineer — Real ML Benchmarks

Amazon Web Services (AWS); Austin (TX); On-site; USD 143,700 - 194,400; Full time

Benefits offered by this job

Sign-on payments

Restricted stock units (RSUs)

Health insurance

401(k) matching

Paid time off

Parental leave

Job summary

Annapurna Labs, an AWS organization, seeks a Post‑Silicon Validation engineer to quantify ML training chip performance against targets. You will bridge silicon capabilities and real‑world workloads—ensuring accelerators deliver on latency, throughput, and efficiency at cloud scale.

You will measure, profile, and optimize across micro-architectures and full model runs, building dashboards and feeding findings back to design teams for production readiness.

Qualifications

• 3+ years of non-internship professional software development experience.
• 2+ years of non-internship design or architecture experience.
• Experience with Machine Learning fundamentals, including training/inference lifecycles, or PyTorch/JAX.
• Bachelor's degree in CS/engineering/math or equivalent; or experience in Java/C++/Python.
• 3+ years of hardware performance counters and profiling tools.
• Strong understanding of memory hierarchies, compute pipelines, and interconnect topologies.
• Experience applying statistical methods and data visualization to performance data.

Responsibilities

• Design and execute performance benchmarks spanning micro-architectures to full model training.
• Measure and analyze compute throughput, memory bandwidth, interconnect latency, and more.
• Profile real ML workloads (transformer models, LLMs, vision models) on silicon.
• Identify performance bottlenecks and work with architecture teams on optimization.
• Build automated performance regression dashboards and tracking infrastructure.
• Correlate silicon measurements against RTL simulation and emulation predictions.

Skills

Software development experience

System design / architecture

Machine Learning fundamentals

Python / Java / C++

Hardware performance counters & prof.

Computer architecture fundamentals

Statistics & data visualization

CUDA kernels / ML kernels

GPU/AI accelerator experience

AllReduce / AllGather knowledge

Education

Bachelor's degree in computer science, engineering, mathematics or equivalent

Tools

CUDA kernels

Job description

Annapurna Labs, an AWS organization, seeks a Post‑Silicon Validation engineer to quantify ML training chip performance against targets. You will bridge silicon capabilities and real‑world workloads—ensuring accelerators deliver on latency, throughput, and efficiency at cloud scale.

You will measure, profile, and optimize across micro-architectures and full model runs, building dashboards and feeding findings back to design teams for production readiness.