Graph Compiler Engineer

NextSilicon is a swiftly growing unicorn startup that is reimagining high-performance computing. Our pioneering coprocessor vastly accelerates supercomputers, driving them forward into a new generation. Our new software-defined hardware architecture enables HPC and AI to fulfill its promise of breakthroughs in all fields of advanced research.

We are looking for a strong software developer with a can-do attitude, and an ability to closely interact with other teams and engineers, driving collaborative efforts to get things done. You will be developing our projection algorithm used in the compilation process of applications to our proprietary compute hardware. The team’s role is to bring our projection code to production quality, allowing a fast and transparent compilation of HPC and AI applications.

Requirements:

BSc or above in computer science or software engineering.
Minimum of 5 years of experience in Python software development in algorithmic projects involving graph, clustering or optimization algorithms.
Experience with C++: an advantage.
Familiarity with compiler engineering such as IR, optimization techniques, dataflow analysis: an advantage.
Experience with pre-silicon SW development and HW/SW interfaces: an advantage.

Responsibilities:

Implement software features and enable hardware features in software.
Analyze and optimize code for improved performance.
Collaborate with NextSilicon researchers on the development of projection algorithms.
Develop and implement tools for debugging, analysis, testing, and logging.