M

MemryX

Edge AI processing accelerator chips

System Hardware Engineer

Hardware Systems IntegrationConfirmed live in the last 24 hours
Full-timeZhubei City, TaiwanIn-person

About the Job

Requirements

  • A bachelor's or master's degree in Electrical Engineering, Computer Engineering, or a related field.
  • At least 5 years of experience in hardware system, motherboard, add-in card, or module design.
  • Solid knowledge of PCIe and M.2 interface standards and design guidelines.
  • Familiarity with SPI, I2C, SMBus, and UART interfaces.
  • Proficiency with EDA schematic and design tools, including OrCAD and Allegro.
  • Hands-on expertise with oscilloscopes and logic analyzers for signal integrity validation and debugging.
  • Strong analytical, problem-solving, and root-cause analysis skills, with cross-functional communication abilities.

Responsibilities

  • Develop hardware products based on AI accelerators, primarily PCIe add-in cards and M.2 modules.
  • Participate in product specification definition, system architecture planning, component selection, and design feasibility evaluation.
  • Perform schematic design and design reviews, and coordinate with PCB layout engineers.
  • Design and validate high-speed interfaces, power circuits, clocks, reset circuits, boot-related circuits, management interfaces, and peripheral circuitry.
  • Perform new board bring-up, functional verification, and system integration testing.
  • Develop hardware validation plans, test procedures, and validation reports.
  • Conduct power, signal integrity, stability, performance, thermal, and long-duration reliability testing.
  • Analyze and resolve issues involving hardware, firmware, drivers, and overall system integration.
  • Provide hardware platforms, test environments, tools, and debugging support for the firmware development team.
  • Assist with debugging issues related to boot flow, PCIe enumeration, I2C/SMBus, SPI, UART, GPIO, reset, and power sequencing.
  • Use oscilloscopes, logic analyzers, and other tools to identify and isolate system-level issues.
  • Work with firmware and IC design teams to perform root-cause analysis and implement corrective actions.
  • Support product validation and issue resolution throughout EVT, DVT, and PVT stages.
  • Assist in developing test fixtures, production test procedures, and hardware diagnostic tools.
  • Work with PCB vendors, EMS/ODM partners, component suppliers, and manufacturing teams to resolve prototype and production issues.
  • Support BOM management, alternative component qualification, ECO/ECN implementation, and product revision control.
  • Support customers during product evaluation, system integration, and deployment.
  • Analyze customer issues related to PCIe compatibility, system stability, performance, power, and thermal behavior.
  • Reproduce customer-reported issues, collect and analyze debug logs, and work with internal engineering teams to provide solutions.
  • Write hardware specifications, validation reports, and customer-facing collateral such as Hardware Design Guides and Application Notes.

Desired Qualifications

  • Experience developing AI accelerator, GPU, NPU, FPGA, PCIe add-in card, or M.2 module products.
  • Familiarity with PCIe Gen3, Gen4, or Gen5, including link training, LTSSM, enumeration, and compatibility debugging.
  • Experience with signal integrity or power integrity analysis.
  • Experience in thermal design, power consumption analysis, or thermal validation.
  • Familiarity with Linux environments and common debugging tools or commands.
  • Experience with server platforms, embedded systems, edge AI devices, or data center hardware.
  • Knowledge of product safety, EMI/EMC, reliability, or compliance testing.
  • Experience in customer technical support, FAE responsibilities, or on-site issue resolution.
  • Ability to communicate technical information in English with overseas customers, suppliers, or business partners.

About the Company

MemryX

Edge AI processing accelerator chips

MemryX makes AI accelerator chips for edge AI. Its MX3 chip uses a configurable native dataflow architecture and in-memory computing to run AI models without needing manual tuning or retraining, and it can be scaled by adding chiplets to increase performance or reduce latency. The system minimizes data movement, allowing a single software stack to run multiple models across multiple chips efficiently. A distinguishing feature is its DRAM-like integration that simplifies deployment and its ability to support new hardware and AI models with the same software, plus a PC-ready AI accelerator card demonstrated in Lenovo ThinkCentre PCs. MemryX’s goal is to expand edge AI capabilities by providing a scalable, easy-to-use hardware-software stack that lowers barriers to deploying AI at the edge and into consumer/enterprise PCs by selling MX3 chips.

Company size
11-50
Company stage
Series B
Total funding
$52M
Headquarters
Ann Arbor, MI
Founded
2019

Recently Posted Jobs

Explore more opportunities at MemryX

See more jobs