Senior Inference ML API SDET

About the role

Cerebras Systems builds the world's largest AI chip, which delivers training and inference speeds over 10 times faster than GPU-based cloud services. The ML API Quality team ensures the reliability of every production release shipped to Cerebras Inference Cloud, working with platform, infrastructure, ML systems and product engineering teams. As a Senior Software Engineer in Test for the ML API features team, you will lead testing strategy and execution for AI/ML models, evaluate accuracy, fairness and performance at scale, and validate software and hardware components for Cerebras API Features.

Key responsibilities

• Architect and own end-to-end test strategies for new features, building scalable tests, frameworks and tooling.
• Lead contributions to industry-standard benchmarks and promote rigorous evaluation methodologies.
• Drive automation initiatives to improve engineering efficiency and test coverage, and make coverage and risk-based testing decisions.
• Mentor junior SDETs on testing methodology, debugging and automation, and advocate for world-class product quality.
• Identify systemic quality gaps and lead cross-functional initiatives to fix them.
• Facilitate technical communication across teams and time zones.

Required profile

• 5+ years of experience in software integration, development or quality engineering.
• Proven experience testing compute, machine learning, networking or storage systems in large-scale enterprise environments.
• Strong debugging track record across distributed, scaled-out deployments.
• Experience leading cross-functional quality initiatives and mentoring engineers across geographically dispersed teams.
• Preferred: hands-on ML workloads (LLM and/or multimodal training or inference), hardware architecture knowledge, microservices deployment and orchestration experience, and prior ownership of a quality engineering culture or test infrastructure.

Required skills

• Deep expertise in automation and programming in Python, C++ or Go, with reusable test frameworks.
• Excellent verbal and written communication, able to present technical findings to engineering and leadership.

The role is hybrid, with in-office presence 3 days per week, based in Sunnyvale, CA or Toronto, Canada.