Amazon
Software Engineer - Hardware Abstraction Layer, AWS Machine Learning Accelerators
$144K–$224K
Austin, TX +1 · Full-time · Posted Sep 17, 2026
- Location
- Austin, TX
Cupertino, CA - Experience
- 2+ years writing functional or performance models of hardware; preferred: 2+ years of full software development life cycle experience.
About the job
Develop and maintain hardware abstraction layer software for AWS machine-learning SoCs, working across hardware and system software teams to support chip functionality, performance, and security.
What you'll do
- Build HALs for newly developed SoC IPs and address system-level architectural issues and debugging.
- Test and deploy software for internal customers; refactor and maintain code throughout the device lifecycle.
- Develop customer tooling and abstractions that make SoCs easier to use.
What they ask for
- Experience programming in languages such as C/C++, Python, Java, or Perl.
- 2+ years writing functional or performance models of hardware such as SoCs, ASICs, GPUs, CPUs, or IP blocks.
- Familiarity with SoC, CPU, GPU, and/or ASIC architecture and micro-architecture.
Nice to have
- 2+ years of full software development life cycle experience.
- Experience with DV teams, SystemC or TLM modeling, model correlation, performance modeling, Modern C++, test frameworks, or multi-threaded simulation.
Pay
165,200.00 - 223,600.00 USD annually (base salary); 143,700.00 - 194,400.00 USD annually (base salary)
Tools
C/C++, Python, Java, Perl, SystemC, TLM, Modern C++ (20 and beyond), PyTest
AI focus
Develop HALs for components used in machine learning systems, such as PCIe and HBM; the role does not involve doing ML, and no ML background is needed.
Model-written summary · the original posting is authoritative