$250,000 - $600,000 USD yearly
Originally posted July 22, 2026 by the employer.
About the role
MatX is seeking a Senior SoC Power Analysis & Optimization Engineer to deliver power-efficient silicon across compute, memory management, and high-speed connectivity. The role focuses on developing power estimation methodologies, calibrating energy models, and driving dynamic and static power reduction for AI accelerator SoCs.
What you'll do
- Develop and maintain power estimation methodologies and build calibrated energy models for early architectural exploration and target definition.
- Perform RTL, gate-level, and post-layout power analysis across SoC subsystems including compute, memory, interconnect, PCIe, HBM, and clock/reset networks.
- Analyze AI workloads and application use cases to identify opportunities for dynamic and static power reduction.
- Collaborate with architects, RTL designers, and physical design teams to evaluate implementation trade-offs and drive power-efficient microarchitecture improvements.
- Define representative workloads and sign-off scenarios to improve power analysis coverage.
- Develop automation, scripts, dashboards, and heuristics to support power analysis and root-cause investigations.
What you'll need
- BS or MS in Electrical Engineering, Computer Engineering, or a related field.
- 8+ years of industry experience in SoC, AI accelerator, GPU, networking ASIC, or high-performance digital design.
- Solid understanding of computer architecture and SoC design.
- Hands-on experience with ASIC low-power design techniques.
- Experience performing RTL and gate-level power analysis.
- Proficiency with Python, Perl, or similar scripting languages.
Nice to have
- RTL development experience using Verilog or SystemVerilog.
- Experience with clock distribution, multiple clock domains, clock gating, and power-aware RTL design.
- Experience with synthesis, static timing analysis, place-and-route, and implementation flows.
- Experience using power analysis tools such as Joules, PowerArtist, or PrimeTime PX (PTPX/PPRTL).
- Experience developing power models, budgeting methodologies, and projection flows.
- Experience with power management techniques including clock gating, power gating, and dynamic power management.
- Experience correlating pre-silicon power estimates with silicon measurements.
- Experience with AI workload characterization and power-performance analysis.
Skills: power estimation methodologies, RTL power analysis, gate-level power analysis, low-power design techniques, power optimization
More semiconductor jobs: Physical design jobs
First seen in our index 87 days ago, well above the 61-day median for Physical Design roles.
Physical Design · Physical Design
|
Open roles in category
780
|
Median days since first seen
61 d
|
Category median salary (all employers)
$187k
|
See the full market breakdown ▾Category comparison, skills in demand, and who else is hiring
| Metric | MatX | All employers we track in this specialty (780 roles · 97 employers) |
|---|---|---|
| Open roles in this specialty | 2 | 780 |
| Open roles in the wider Physical Implementation family | 4 | 1750 · 129 employers |
| Median days since first seen | 87 d | 61 d (−26 d vs this employer) |
| Median salary (USD postings) | n/a | $187k |
Skills observed across this category: power estimation methodologies, RTL power analysis, gate-level power analysis, low-power design techniques, power optimization
Who's hiring in this category
- Qualcomm · 121 open roles · median 68 d
- AMD · 67 open roles · median 46 d
- Broadcom · 53 open roles · median 26 d
- NVIDIA · 46 open roles · median 87 d
- Synopsys · 43 open roles · median 45 d
- Apple · 32 open roles · median 10 d
How we counted: 780 open Physical Design (Physical Design) roles from 97 employers tracked in the SemiconductorJobs index, counted 16 Nov 2026. Specialty figures count only roles carrying this exact specialty label, so an employer's related work in neighbouring specialties is not included there, and is counted in the wider Physical Implementation family row. Days are counted from when our index first saw each posting, not from the employer's original posting date.