Senior Math Libraries Engineer - LLM Integration and Developer Experience
Overview:
In this role you accelerate API integration across the CUDA-X math library stack for modern LLM and AI workflows. You join a cross-functional team developing GPU-accelerated libraries and SDKs that power AI and scientific computing. You will architect and implement LLM-friendly APIs and documentation, collaborate on timely library releases, and build tooling to measure AI-generated code quality. You will prototype representations of math APIs for code generation tools, positively impacting developer experiences at scale. This position offers the chance to shape high-performance AI tooling at a world-leading GPU computing company.
Compensation / Benefitsequitybenefits
Responsibilities:
Architect and implement LLM-first math library APIs and documentation structuresCollaborate with internal/external stakeholders to deliver LLM-enhanced library releasesDevelop metrics, tools, and processes to assess impact and quality of LLM/agentic code generationPrototype tooling to convert math library APIs into LLM-friendly representations across popular codegen tools
Key requirements:
PhD or MSc in Computational Science, Computer Science, Applied Math, or related field preferred (or equivalent experience)3+ years of experienceStrong knowledge of LLMs, finetuning, RL, building RAGs, MCP, and agenting toolingProven experience designing clear, composable APIs and writing high-quality, well-documented codeAbility to prioritize multiple projects and work independentlyExcellent collaboration, communication, and documentation habitscollaborationcommunicationdocumentation habitsLLMs and AI toolingAPI design and documentationCUDA-X libraries exposure (cuBLAS, cuFFT, cuSOLVER, cuSPARSE)