C/C++ Hardware / Software Co-Design SDE, Machine Learning Acceleration Systems
Amazon · Austin, TX · Software Development
About this role
Amazon is hiring a mid-level Embedded Software Engineer in the software engineering function based in Austin, TX. The posting calls out experience with Java, C++, C, C#. Compensation is listed at $165,200–$223,600 per year.
- Role
- Embedded Software Engineer
- Function
- software engineering
- Level
- mid
- Track
- Individual contributor
- Employment
- Full-time
- Location
- Austin, TX
- Department
- Software Development
- Posted
- May 5, 2026
More roles at Amazon
Job description
from Amazon careersAnnapurna Labs stands at the forefront of hardware/software co-design, leading innovation not just within Amazon Web Services (AWS) but across the entire industry. We design and build every component of our hardware and software to create best-in-class machine learning products that accelerate customer workloads through industry leading hardware using our custom silicon solutions. If you're passionate about building the highest-performing, hardware-accelerated Machine Learning systems and want to be part of the entire journey from pre- through post-silicon development, Annapurna Labs offers an exceptional career opportunity. Join us in shaping the future of AI acceleration At the heart of AWS Machine Learning servers reside our custom Amazon-designed silicon that powers next-generation machine learning capabilities for our customers. We're seeking an experienced C/C++ engineer to join our embedded software team, where you'll develop bare metal firmware that drives neural network model execution across our custom ASIC-based ML Accelerator chips. Working at the intersection of hardware and software, you'll collaborate closely with our architecture and design teams to co-develop the firmware and custom hardware that enables machine learning within our accelerator chips. Our mission is ambitious: to democratize access to industry-leading ML infrastructure and make deep learning capabilities accessible to everyday software developers.…