AI GPU Performance Engineer, Modular

$157.5-253k

CUDA

Mid and Senior level

Remote from Canada, US

Fast, scalable Gen AI inference platform

Job no longer available

Fast, scalable Gen AI inference platform

201-500 employees

B2BArtificial IntelligenceMachine Learning

Job no longer available

$157.5-253k

CUDA

Mid and Senior level

Remote from Canada, US

201-500 employees

B2BArtificial IntelligenceMachine Learning

Company mission

To have real, positive impact in the world by reinventing the way AI technology is developed and deployed into production with a next-generation developer platform.

Job

Company

Role

Who you are

3+ years of relevant experience working on complex code and systems
NOTE: This role is also open to candidates with extensive experience beyond the specified 3 year minimum
Experience with GPU programming languages such as CUDA, OpenCL, SYCL, SPRIV, or C++AMP
In-depth understanding of recent GPU architectures and performance optimizations
Experience with AI workloads and performance tuning considerations such as fusion strategies
Experience with core AI kernels such as matrix multiply and convolution
Strong familiarity with using GPU profilers, debuggers, etc
Deep interest in machine learning technologies and use cases
Creativity and curiosity for solving complex problems, a team-oriented attitude that enables you to work well with others, and alignment with our culture

What the job involves

As an AI GPU Performance Engineer, you will own architecting performance libraries for GPUs. You will develop kernels and algorithms to increase performance of kernels, reduce the activation volumes, speedup data pre- and post-processing, and improve the end-to-end performance of the AI workloads
Design, develop, and optimize high-performance AI numeric and data manipulation kernels/operators for GPUs
Achieve state-of-the-art performance by leveraging software and micro-architectural features of GPUs
Work with compiler, framework, and runtime teams to deliver end-to-end performance that fully utilizes GPU workstations and servers
Collaborate with machine learning researchers to guide system development for future ML trends

Our take

AI development is booming, but the tools behind it? Not so much. Developers are often stuck dealing with messy, disconnected systems that slow everything down and bump up costs.

To address this, Modular has developed a unified AI platform that streamlines the development process. It's AI Engine enhances the performance of models on CPUs and GPUs, supporting popular frameworks like TensorFlow and PyTorch. And on top of that, it's built Mojo, a programming language that's as easy as Python but much faster (35,000 times faster).

In 2023, the company raised an impressive $100M to power its mission. It's been using this funding to grow its team, support more hardware, and push Mojo even further. With the hopes of making AI development faster, smoother, and a whole lot more fun for everyone.

Kirsty

Company Specialist at Welcome to the Jungle

Company

Funding (last 2 of 3 rounds)

Sep 2025

$250m

LATE VC

Aug 2023

$100m

LATE VC

Total funding: $380m

Company benefits

A variety of fantastic health benefits (health, dental, vision insurance; life insurance etc) are available
A 401k plan with up to 5% match
Free tax advice on Carta
Generous work-from-home stipend of $1500 to help you improve your home office
Unlimited paid time off and flexible work hours

Leadership

Chris Lattner

(Co-Founder & CEO)

Founder, Architect, Engineer and BDFL for the LLVM Organization. Former President of Engineering and Product at SiFive, as well as working in various roles at Google, Tesla, and Apple.

Tim Davis

(Co-Founder & President)

Worked at Google for 5 years, ultimately as Group Product Lead for Google ML. Previously founded Fluc Inc, where they developed the entire ML powered delivery system with their co-founder, and mapping software system CrowdSend.

Share this job

View 13 more jobs at Modular