About Rain:

Our mission is to radically reduce the cost of artificial intelligence.

We are the world leaders in algorithm/hardware co-design for artificial intelligence. Our roadmap begins with products 100x better than GPUs and will ultimately deliver products that are many orders of magnitude more cost effective than what is available today. We will ultimately be able to put models the size of ChatGPT into chips the size of a thumbnail.

About the role:

The AI HW/SW Co-Design Intern will contribute to developing a framework for an in-memory compute AI accelerator to perform hardware/software co-design and optimization. This framework will guide the architecture and algorithm design to optimize Rain’s hardware. This is a collaborative role – as an intern, you will have the opportunity to work across many teams at Rain.

This is a remote internship opportunity – you can work anywhere in the US. The duration of this internship will be 6 months.

Responsibilities:

Lead and contribute to the HW/SW co-optimization of Rain’s AI accelerators, including but not limited to:

• Create functional models for Rain’s chips.
• Develop an HW/SW codesign framework.
• Optimize hardware architecture for Rain’s benchmarking suite.
• Develop new models for Rain’s next-generation hardware.
• Collaborate with the algorithms team to develop hardware-aware algorithms.

Document and present results.

Required Qualifications:

• Currently pursuing a PhD in Computer Engineering, electrical engineering or a related field.
• Strong Knowledge of computer architecture and AI accelerator architecture.
• Experience with hardware models for accelerators and RISC-V processors.
• Experience with DNN models mapping to hardware.
• Experience with hardware-software co-design and design space exploration.
• Experience in developing and debugging in C/C++, Python and/or PyTorch.
• Excellent communication skills.

Preferred Qualifications:

• Publications in top conferences on computer architecture and design automation conferences or related topics.

This is position will be paid hourly. The anticipated hourly rate for this role is $45 to $85/hour.