Senior Math Libraries Engineer, LLM Integration and Developer Experience

What you'll do

• Drive the architecture and implementation of math library APIs and documentation structures designed for LLM-first use.
• Ensure math library APIs are introspectable, explainable, and easily synthesized by AI agents.
• Deliver timely LLM-enhanced library releases in coordination with internal and external stakeholders.
• Build metrics, tools, and processes to measure the impact and quality of LLM-generated code.
• Prototype solutions to transform math library APIs into LLM-friendly representations across popular codegen tools.
• Design clear, composable APIs and write high-quality code for complex technical domains.

What we're looking for

• PhD or MSc degree in Computational Science, Computer Science, Applied Math, or a related field (preferred).
• 3+ years of experience.
• Robust knowledge of LLMs, finetuning, RL, building RAGs, MCP, and building agentic tooling.
• Proven experience designing clear, composable APIs and writing high-quality, well-documented code for complex technical domains.
• Ability to prioritize multiple projects and work independently with minimal direction.
• Excellent collaboration, communication, and documentation habits.
• Prior work in AI-assisted software engineering, code generation, or programming language design (preferred).
• Familiarity with CUDA-X math library APIs such as cuBLAS, cuFFT, cuSOLVER, and cuSPARSE (preferred).