Groq is seeking a Deep Learning Performance Specialist for a full-time position located in Menlo Park, CA.


Groq is a machine learning systems company building easy-to-use solutions for accelerating artificial intelligence workloads. Our work spans hardware, software, and machine learning technology. We are seeking exceptional software engineers to join our compiler team in tackling the exciting problem of translating ML programs written in existing ML frameworks into the language of our custom Groq ML accelerator chips.

We are looking for deep learning specialists to be based in either our Menlo Park, CA or Portland, OR office.

ESSENTIAL DUTIES & RESPONSIBILITIES

Optimize model performance: Leverage your knowledge of machine learning workloads to identify compiler optimizations that maximize program execution efficiency on Groq hardware platforms. Design compiler features: Work with members of the Groq’s compiler team to design and evaluate potential features for our compiler. Collaborate closely with hardware designers to identify ways to take advantage of hardware capabilities. Manage performance tradeoffs: Understand the balance between performance, memory, and power in compiler generated code. Innovate: Contribute to the design of next generation hardware architectures and compiler releases. Collaborate with others through code reviews, design discussions, and problem-solving sessions.

SKILLS & QUALIFICATIONS Minimum qualifications: Master’s degree in computer science or mathematics, or equivalent practical experience 3+ years experience developing and optimizing deep learning models Strong coding skills Experience with TensorFlow and other common frameworks (ONNX, Caffe, PyTorch) Excellent collaboration skills, with strong verbal and written communication

Preferred qualifications: PhD in computer science or mathematics Familiarity with functional programming languages (such as Haskell) Understanding of compiler architecture and compiler implementation techniques

Keywords: Haskell, TensorFlow, machine learning, deep learning, compiler, functional programming, hardware, accelerator, software engineer, startup

To Apply

Send resume to apply-9149 [at] functionaljobs [dot] com