Job Search and Career Advice Platform

Enable job alerts via email!

High-Performance Kernel Engineer for AI Hardware

Cerebras

United Arab Emirates

On-site

AED 367,000 - 478,000

Full time

3 days ago
Be an early applicant

Generate a tailored resume in minutes

Land an interview and earn more. Learn more

Job summary

A cutting-edge AI technology firm is seeking a Kernel Engineer to develop high-performance software solutions for AI workloads. You will design, optimize, and scale deep learning operations, and work on algorithms to leverage the capabilities of custom hardware architecture. Required qualifications include a degree in Computer Science or related fields, along with proficiency in C++ and Python. Join a world-class team pushing the limits of AI innovation.

Benefits

Open-source AI research opportunity
Work on fast AI supercomputers
Inclusive work culture

Qualifications

  • Bachelor's, Master's, PhD or foreign equivalents in relevant fields.
  • Understanding of hardware architecture concepts.
  • Skilled in C++ and Python programming.

Responsibilities

  • Develop design specifications for machine learning kernels.
  • Debug kernel library for optimized low-level routines.
  • Measure software performance to inform design decisions.
  • Integrate testing methodologies to verify kernel libraries.

Skills

C++ programming
Python programming
Debugging complex software
Understanding hardware architecture

Education

Bachelor’s, Master’s, PhD in Computer Science, Computer Engineering, Mathematics
Job description
A cutting-edge AI technology firm is seeking a Kernel Engineer to develop high-performance software solutions for AI workloads. You will design, optimize, and scale deep learning operations, and work on algorithms to leverage the capabilities of custom hardware architecture. Required qualifications include a degree in Computer Science or related fields, along with proficiency in C++ and Python. Join a world-class team pushing the limits of AI innovation.
Get your free, confidential resume review.
or drag and drop a PDF, DOC, DOCX, ODT, or PAGES file up to 5MB.