GlobalFoundries is looking for students with experience in hardware accelerator architectures to engage in advanced system-level analysis of Neural Processing Unit (NPU) architectures.
Requirements
- solid foundation in digital design principles, including memory systems and parallel processing.
- Familiarity with RTL design using Verilog or VHDL
- experience with synthesis flows and physical design tools
- proficiency in scripting languages such as Python or TCL for automation and analysis
Responsibilities
- Conduct in-depth architectural analysis of NPUs, emphasizing dataflow optimization and memory hierarchy.
- Investigate the performance and efficiency of systolic array configurations and their integration with buffer memory systems.
- Perform RTL synthesis and Place-and-Route (PnR) using industry-standard EDA tools.
- Analyze Power, Performance, and Area (PPA) metrics to guide design decisions and architectural trade-offs.
Other
- Actively pursuing a Master’s or Ph.D in engineering or related field through an accredited degree program during the time of internship.
- Must have at least an overall 3.0 GPA and be in good academic standing.
- Language Fluency - English (Written & Verbal)
- Ability to work at least 40 hours per week during the 3-4 month internship.
- Prior related internship or co-op experience
