Failure Analysis Engineer, Electrical Engineering

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

Etched is hiring a Failure Analysis Engineer to own the end-to-end debug process across our full hardware stack: chip, board, and rack-scale systems.

You will be responsible for rapidly diagnosing, triaging, and resolving hardware failures; determining whether issues originate in the chip, board, or rack infrastructure; and driving resolution with the appropriate team. This is a highly cross-functional role, working closely with US-based hardware and silicon teams to build and refine debug playbooks as production scales.

The ideal candidate has deep EE fundamentals, systems-level debugging experience, and the ability to solve hard problems under pressure.

Key responsibilities

• Own failure triage across the stack. Receive field and production failures, isolate whether the root cause is chip, board-level, or system/rack-level, and route to the appropriate team with a clear problem statement.

• Drive root cause analysis using electrical test equipment (oscilloscopes, logic analyzers, multimeters) and system-level diagnostics to identify failure mechanisms and determine corrective actions.

• Build and refine debug processes. Partner with US hardware counterparts to document debug flows for different failure modes, creating repeatable playbooks that scale with production volume.

• Debug rack-level issues. Troubleshoot communication failures between rack managers, CDUs, and system components. Understand how thermal, power, and network infrastructure interact at the rack scale.

• Interface with BMC and system firmware. Use Linux command line and BMC interfaces to pull logs, run diagnostics, and validate system health during failure investigations.

• Close the loop on quality. Feed failure trends and root cause findings back to design, manufacturing, and operations teams to drive systemic improvements.

You May Be a Good Fit If You Have

• Bachelor’s or Master’s degree in Electrical Engineering or a related field.

• Fluency in oscilloscopes, signal integrity basics, power delivery, and board-level debug.

• Systems-level thinking. Strong understanding of how servers work end-to-end: BMC, BIOS, OS, thermals, and power sequencing. Can debug issues that span multiple subsystems.

• Linux command line proficiency. Comfortable with CI pulling logs, running scripts, and navigating server environments from the terminal.

• Strong communication skills across teams. You can translate a complex hardware failure into a clear problem statement for silicon, firmware, or mechanical teams. You've worked across time zones and functions.

• Composure under pressure. Production failures don't wait. You're energized by urgent, ambiguous problems and take ownership until they're resolved.

• 3+ years of experience in hardware debug, failure analysis, or systems engineering in a server, datacenter, or semiconductor environment.

Nice to Haves

• Rack-scale infrastructure (cooling systems, power distribution, rack managers)

• High-speed interfaces (PCIe, Ethernet, SerDes) and their common failure modes

• ATE or production test environments

• Experience with Datacenters, GPUs, FPGAs, or custom ASICs

• We encourage you to apply even if you do not believe you meet every single qualification.

Benefits

• Competitive compensation packages, including generous equity packages

• Comprehensive insurance coverage and other top-of-market benefit

• US onboarding experience in our Silicon Valley headquarters

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.