Cerebras Systems has pioneered a groundbreaking chip and system that revolutionizes deep learning applications. Our system empowers ML researchers to achieve unprecedented speeds in training and inference workloads, propelling AI innovation to new horizons.

The Condor Galaxy 1 (CG-1), unveiled in a recent announcement, stands as a testament to Cerebras' commitment to pushing the boundaries of AI computing. With a staggering 4 ExaFLOP processing power, 54 million cores, and 64-node architecture, the CG-1 is the first of nine powerful supercomputers to be built and operated through an exclusive partnership between Cerebras and G42. This strategic collaboration aims to redefine the possibilities of AI by creating a network of interconnected supercomputers that will collectively deliver a mind-boggling 36 ExaFLOPS of AI compute power upon completion in 2024.

Cerebras is building a team of exceptional people to work together on big problems. Join us!

About The Role
As a Runtime Engineer, you will directly impact the performance at which deep learning models are trained on our “distributed systems” hardware and be responsible for enabling next-generation AI applications that require substantial computational capabilities. In this position, you will develop algorithms for execution, acceleration, partitioning, and routing of communication for dataflow graphs on a massively parallel, multi-core architecture.

Specific responsibilities may include:

• Be able to understand the flow of data in a distributed system and how to characterize performance pain points
• Develop algorithms for allocation of compute, communication, and memory resources
• Measure, analyze, and improve execution of Runtime software (that is responsible for training large models with massive datasets)
• Integrate successful optimizations into production software stack
• Implement mathematical models in C++ or Python using discrete optimization techniques and standard libraries and packages

Requirements

• Currently enrolled in a University in Computer Science, Computer Engineering, or any other related discipline
• Strong proficiency in C/C++
• Familiarity with Python or other scripting language
• The ability to operate at multiple levels of abstraction in the software stack

Preferred

• Knowledge about distributed systems, memory subsystem of modern computers, and networking solutions

Term Length

• 12-16 months starting May 2024

Please apply to the job with BOTH your resume and transcript (official or unofficial).