Annapurna Labs

Technology

MLAcceleratorPerformanceValidationEngineer,PostSiliconValidation

$70–194k Austin, Texas, United States FULL TIME
Market Sentiment
HIGH DEMAND

Neural analysis suggests this role is
optimal for Mid+ candidates.

The Brief

“ML Accelerator Performance Validation Engineer, Post Silicon Validation at Annapurna Labs. Skills: ML hardware, Performance validation, Post-silicon validation, ML workloads. Design performance benchmarks. Execute performance benchmarks”

What You'll Achieve.

Ensure accelerators deliver promises; Determine silicon readiness

Industry & Context.

Technology
Problems you'll solve

Root cause analysis; Troubleshooting

What They're Looking For.

Must Have

3+ years professional software development, 2+ years system design or architecture, Machine Learning fundamentals, Large Language Model fundamentals, PyTorch or JAX software experience, Bachelor's degree in CS, engineering, math, 3+ years hardware performance counters, 3+ years profiling tools experience, Computer architecture fundamentals understanding, Statistical methods application, Regression analysis application, Data visualization techniques application

Nice to Have

3+ years full software development lifecycle, CUDA kernels experience, ML/low-level kernels experience, Developing LLMs in production, Deploying LLMs in production, Knowledge of collective communications, Experience with HBM, Experience with PCIe, Experience with DMA bandwidth characterization

What You'll Do.

Design performance benchmarks

Execute performance benchmarks

Measure compute throughput

Analyze compute throughput

Measure memory bandwidth

Analyze memory bandwidth

Measure interconnect latency

Analyze interconnect latency

Identify performance bottlenecks

Build automated dashboards

Track performance infrastructure

Correlate silicon measurements

Develop training models

How You'll Work.

Team & Collaboration

Architecture teams; Design teams; Cross-functional teams

Full Job Description

Annapurna Labs, an AWS organization with development centers in the U.S. and Israel, builds custom silicon and software for AWS customers. Our team combines cloud-scale innovation with world-class expertise across silicon engineering, hardware design, verification, software, and operations to tackle technical challenges that have never been seen before. Join our Post-Silicon Validation team to quantify and qualify the performance of AWS's custom ML training chips against architectural targets. You'll bridge the gap between silicon capabilities and real-world ML workload demands — ensuring our accelerators deliver on latency, throughput, and efficiency promises at cloud scale. You'll work in a fast-paced, startup-like environment alongside some of the brightest minds in the industry on next generation AI/ML hardware that powers AWS's training and inference infrastructure. Your analysis will directly shape architectural decisions for next-generation accelerators and determine when silicon is ready for production deployment. Key job 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 A day in the life Your primary focus is measuring and understanding how our AI chips perform under real workloads. You'll spend mornings digging into benchmark results — figuring out where cycles are being lost and why throughput isn't hitting targets. When something looks off, you'll instrument the hardware, profile the pipeline, and work with design teams to get it fixed. Some days you'll be developing and running full training m

Free ATS check

Applying for this ML Accelerator Performance Validation Engineer, Post Silicon Validation role?

Most applicants get filtered before a human reads their resume. See if yours makes the cut.

ANONYMOUS · UNFILTERED

What do employees actually say about Annapurna Labs?

Real rants from real employees. Read before you apply.

Read Company Rants →