About Mindbeam
We are building the next-generation AI infrastructure for open source and enterprise. Our work is deeply research-oriented and passionate about developing ground-breaking innovations to take state-of-the-art AI applications to the next level.
Mission
Push the boundaries of performance by developing custom kernels and low-level optimizations for next-generation AI workloads.
Role Expectations
• Design and implement custom GPU/accelerator kernels to maximize performance.
• Profile, benchmark, and optimize critical ML workloads.
• Collaborate with researchers to translate algorithmic advances into efficient, production-ready code.
• Stay current with hardware advancements (CUDA, ROCm, TPU) to inform kernel design.
• Document and share best practices for low-level optimization.
Background
• Bachelor’s, Master’s, or PhD in Computer Science, Electrical Engineering, or related field—or equivalent experience.
• 2+ years of experience in GPU programming, parallel computing, or systems-level optimization.
• Strong coding skills in C++, CUDA, or similar languages.
• Familiarity with ML frameworks and their low-level backends.
• Experience optimizing workloads for distributed and heterogeneous compute environments.
• Comfort with profiling tools and performance diagnostics.
About You
You are detail-oriented, performance-obsessed, and excited by the challenge of squeezing out every ounce of compute efficiency. You enjoy working at the intersection of algorithms and hardware, and you thrive in a collaborative environment where bold ideas are encouraged.
Litespark is a language model framework which utilises advanced algorithms to speedup training and inference workloads for generative AI applications.
Key team members
Oscar Chavez
Nii Osae
Kevin Chao
Ian Abkowitz
Jobr aggregates jobs directly from company career portals — no middlemen. Our team applies on your behalf with AI-tailored resumes, reviewed by a human before submission.