
STEM Computational Scientific Software & Evaluation Design - Astrophysics & Cosmology
Job Description
This position is listed on behalf of a partner company, who manages all applications and next steps. Our partner is looking for a STEM Computational Scientific Software & Evaluation Design - Astrophysics & Cosmology based in Canada.
This role sits at the intersection of astrophysics research, computational science, and AI evaluation design, where you will build advanced scientific problem benchmarks used to test frontier AI systems. You will design graduate-level computational tasks grounded in real astrophysical and cosmological workflows, leveraging scientific libraries and simulation tools. The work focuses on creating challenges that go beyond computation, emphasizing reasoning, experimental design, and data interpretation under realistic scientific constraints. You will iterate on problem difficulty by testing against state-of-the-art models and refining based on calibration feedback. The environment is highly technical, research-driven, and collaborative with experts in AI and scientific computing. This is a unique opportunity to directly shape how next-generation AI systems understand and reason about the universe.
This position is listed on behalf of a partner company, who manages all applications and next steps. Our partner is looking for a STEM Computational Scientific Software & Evaluation Design - Astrophysics & Cosmology based in Canada.
This role sits at the intersection of astrophysics research, computational science, and AI evaluation design, where you will build advanced scientific problem benchmarks used to test frontier AI systems. You will design graduate-level computational tasks grounded in real astrophysical and cosmological workflows, leveraging scientific libraries and simulation tools. The work focuses on creating challenges that go beyond computation, emphasizing reasoning, experimental design, and data interpretation under realistic scientific constraints. You will iterate on problem difficulty by testing against state-of-the-art models and refining based on calibration feedback. The environment is highly technical, research-driven, and collaborative with experts in AI and scientific computing. This is a unique opportunity to directly shape how next-generation AI systems understand and reason about the universe.