Responsibilities
THE ROLE:
As an AMD GFXIP Architecture Engineer
, you will contribute to the architecture of next-generation AMD graphics and AI accelerator products. You will work closely with GPU, AI, SoC, and system architects to define architectural capabilities that improve functionality, scalability, performance, security, and reliability across client, data center, and automotive markets.
A significant focus of this role is designing architectural mechanisms that improve system functionality, robustness, resiliency, diagnosability, and recoverability. Through collaboration with experienced architects across AMD, you will gain broad exposure to GPU architecture, system design tradeoffs, and large-scale silicon development.
We are seeking strong hardware architects with experience in CPUs, SoCs, accelerators, networking, storage, automotive, or other complex hardware systems who are interested in solving challenging architectural problems and growing in new technical areas. This role provides the opportunity to influence real production silicon while working closely with cross-functional teams across the company.
THE PERSON:
You are a technically strong engineer with a passion for computer architecture and complex hardware systems. You enjoy analyzing architectural tradeoffs and designing features that improve how products behave in real-world deployments. You thrive in collaborative environments and enjoy working across disciplines to solve challenging technical problems.
You are comfortable asking questions, learning from experienced engineers, and expanding your expertise into new technical domains.
Prior experience with GPU architecture or RAS technologies is beneficial but not required. Curiosity, architectural thinking, strong engineering fundamentals, and a growth mindset are more important than specific domain expertise.
KEY RESPONSIBILITIES:
- Contribute to the definition and development of architectural features and capabilities for next-generation GFXIP deployments, with a particular focus on system reliability, resiliency, diagnosability, and recoverability.
- Assist in writing and reviewing architectural specifications for new hardware capabilities and architectural enhancements.
- Work with GPU, AI, and SoC architects to develop solutions for system robustness, fault management, recovery, observability, and serviceability requirements.
- Analyze architectural tradeoffs involving performance, power, area, security, and reliability, and help evaluate their impact on product goals.
- Support hardware design teams in evaluating resiliency mechanisms, failure-management strategies, and architectural coverage while identifying opportunities for future improvements.
- Collaborate across architecture, hardware design, software, SoC, and product teams to ensure architectural requirements are understood and successfully implemented.
- Help analyze product reliability metrics such as FIT, DPPM, MTBF, and MTBR, and how architectural decisions influence those outcomes.
- PREFERRED EXPERIENCE: Familiarity with architectural techniques that improve system robustness, error detection, recovery, diagnosability, and resiliency.
- Knowledge of concepts and mechanisms such as ECC, parity, watchdog timers, heartbeat monitors, fault containment, and recoverable error handling.
- Experience with RTL design and/or verification.
- Experience with GPUs, AI accelerators, SoCs, CPUs, networking, storage, automotive, or other large-scale hardware designs.
- Understanding of reliability metrics such as FIT, DPPM, MTBF, MTBR, or related figures of merit.
- EDUCATION
- Bachelor's degree required; Master's or PhD in Computer Engineering, Electrical Engineering, Computer Science, or a related field is a plus.
- LOCATION
- Austin, TX
- This role is not eligible for visa sponsorship.
Qualifications
