Senior Research Engineer - Training Efficiency
Luma AIFull Time
Senior (5 to 8 years)
Candidates must have experience with PyTorch API and implementation, CUDA system design, asynchronous programming models, and heterogeneous compute, with proficiency in C++ and libtorch. Bonus qualifications include machine learning model architecture design, TensorRT, CUDA kernel implementation, and experience with autonomous vehicle multi-sensor model architectures.
The PnP Optimization Engineer will collaborate with research scientists on algorithm design, identify and communicate best practices for model development, and reformulate performance bottlenecks. Responsibilities also include developing a framework for mixed precision training of multi-task, multi-modality models in a distributed environment, researching and implementing alternative formulations for DNN operations and AV centric representations, and enabling safe and efficient deployment of PnP models.
Develops self-driving technology for trucking
Waabi focuses on developing self-driving technology specifically for the trucking industry. Their main product, the Waabi Driver, is designed to learn from experiences, adapt to new situations, and work with various hardware setups. This technology aims for large-scale use and is intended for integration into vehicles by original equipment manufacturers (OEMs). A key feature that sets Waabi apart from competitors is its use of Waabi World, an advanced simulation environment that allows for efficient development of their self-driving systems. This reduces the need for extensive real-world testing, which can be costly and time-consuming, while also improving safety and adaptability. Waabi's goal is to transform the trucking industry by providing technology that lowers operational costs and enhances safety for commercial vehicles. They aim to license their technology to trucking companies and OEMs, making it easier for these businesses to adopt self-driving solutions.