Supercomputing Test Software Engineer

About Etched
Etched is building the world's first AI inference system purpose-built for transformers - delivering over 10x higher performance and dramatically lower cost and latency than a B200. With Etched ASICs, you can build products that would be impossible with GPUs, like real-time video generation models and extremely deep & parallel chain-of-thought reasoning agents. Backed by hundreds of millions from top-tier investors and staffed by leading engineers, Etched is redefining the infrastructure layer for the fastest growing industry in history.

Job Summary
We are seeking highly motivated and detail-oriented Software Engineers to join our Supercomputing Testing team. This team plays a critical role in ensuring the reliability and stability of our highest-performance Inference server hardware and software. As a Software Engineer on this team, you will design, develop, and execute comprehensive supercomputing test suites, analyze test results, and collaborate with hardware and software engineering teams at Etched and our ODM partners to identify and resolve potential issues. You will be at the forefront of ensuring our server products meet the highest quality standards before they reach our customers.

Key Team Responsibilities

• Test Development: Design, develop, and implement automated supercomputing test suites using common scripting languages (Python, Go, Bash) and test frameworks across all aspects of System Operation including: boot sequences, root-of-trust, system management, workload deployment and performance.
• Test Execution: Execute tests on server hardware, monitor system performance and health, and analyze test results.
• Failure Analysis: Investigate and debug hardware and software failures identified during testing, providing detailed reports and mitigation plans.
• Collaboration: Collaborate with internal and external hardware and software engineering teams to identify root causes of failures and implement corrective actions.
• Test Infrastructure: Contribute to the development and maintenance of the supercomputing testing infrastructure, including portable test environments and automation tools runnable in any environment.
• Documentation: Create and maintain comprehensive documentation for test plans, test cases, and test results.
• Performance Analysis: Analyze system performance metrics to identify potential bottlenecks and areas for optimization.
• Continuous Improvement: Participate in continuous improvement efforts to enhance the efficiency and effectiveness of the testing process.

Representative projects

• Develop automated test suites to stress-testing of Supercomputing AI subsystems under extreme at-scale workloads.
• Design and implement fault injection tests to simulate hardware and software failures.
• Create tools to monitor and analyze system performance metrics, such as CPU utilization, cross-socket memory performance and usage, and network latency.
• Build and maintain a scalable testing environment capable of handling multiple server configurations.
• Collaborate with hardware engineers to develop tests for new server features and components.
• Contribute to the creation of Pod-level dashboards tracking system testing across hundreds of AI accelerators.

You may be a good fit if you have

• Proficiency in at least one scripting language (e.g., Python, Bash, Go).
• Experience with software testing methodologies and tools.
• Strong understanding of operating systems (Linux preferred) and server hardware architectures.
• Ability to analyze complex technical problems and provide effective solutions.
• Excellent communication and collaboration skills.
• Ability to work independently and as part of a team.
• Experience with version control systems (e.g., Git).
• Experience with reading and interpreting hardware logs.

Strong Candidates May Also Have Experience With (Nice-to-have Qualifications)

• Experience with hardware burn-in testing or reliability testing.
• Experience with performance testing and benchmarking tools.
• Familiarity with hardware diagnostic tools and techniques.
• Experience with CI/CD pipelines.
• Knowledge of low level hardware communication protocols (i2c, etc.)
• Experience with data analysis tools and techniques.

Benefits

• Competitive compensation packages including generous equity packages
• Comprehensive insurance coverage and other top-of-market benefits

How We're Different
Etched believes in the Bitter Lesson. We think most of the progress in the AI field has come from using more FLOPs to train and run models, and the best way to get more FLOPs is to build model-specific hardware. Larger and larger training runs encourage companies to consolidate around fewer model architectures, which creates a market for single-model ASICs.

We are a fully in-person team in San Jose and Taipei, and greatly value engineering skills. We do not have boundaries between engineering and research, and we expect all of our technical staff to contribute to both as needed.