Helm.ai builds AI software for autonomous driving and robotics, aiming to make autonomous technologies more efficient, safer, and accessible. The business problem is to optimize ML models on GPUs or AI accelerators to achieve these goals.
Requirements
- Proficiency in ML model development and optimization techniques (e.g. numerical optimization, quantization, sparsity, pruning, architecture search and design), particularly on model deployment onto GPU’s or AI accelerators
- Strong understanding of deep learning algorithms, software engineering and GPU-based computing
- Experience working with neural networks in Tensorflow and/or PyTorch
- Proficiency in Python
- Computer vision experience
Responsibilities
- Research, prototype and evaluate state of the art model optimization techniques and algorithms
- Characterize neural network quality and performance based on research, experiment and performance data and profiling
- Incorporate optimizations and model development best practices into existing ML development lifecycle and workflow.
- Define the technical vision and roadmap for DL model optimizations
- Write technical reports indicating qualitative and quantitative results to colleagues and customers
- Develop, deploy and optimize deep learning (DL) models on various GPU and AI accelerator chipsets/platforms
Other
- Proven ability to thrive in fast-paced environment
- Ability to communicate complex technical concepts to colleagues and a variety of audience
- Introspection, thoughtfulness, and detail-orientation
- Master’s or Ph.D. in a related field and/or 5+ years of experience in a directly related field
