We are a venture-backed early-stage startup developing processors specialized for machine learning. The processor will provide orders or magnitude improvement in speed and power efficiency with a goal of unseating the GPU as the dominant computing platform for AI.
We are seeking an expert in AI/ML to serve as a key processor architect. The AI/ML expert is expected to be aware of the latest developments in AI/ML and ensure that the processor provides orders-of-magnitude improvement in processing speed for industry-relevant AI/ML problems. This is an exciting opportunity to join a startup working on an industry disrupting problem at the earliest possible stage.
The ideal candidate has deep understanding of the full-stack of machine learning architectures, insatiable curiosity, is interested in cross-disciplinary problems, and loves to optimize algorithms.
Responsibilities:
Collaborate with hardware engineers to develop an accurate simulation environment of the AI/ML processor
Develop new algorithms to train models in novel processing environments
Train AI/ML models using the simulated processor
Work with hardware engineers to define and refine the processor architecture based on insights learned through simulation
Continuously drive to understand the interplay of physics, computing architectures, and machine learning algorithms
Qualifications:
PhD in representation learning, theory of computation, unsupervised learning, or similar field
Has built and trained neural networks from scratch
Experience developing and training new AI/ML processing architectures
Deep knowledge of the structure and internal operation of neural networks
Curiosity and interest in learning about physical systems and hardware
Fluent knowledge of Python
Fluency in PyTorch, TensorFlow, JAX, or other industry-standard ML software libraries