Riley Dulin is a seasoned software engineer at Meta, where he has honed his expertise in systems-level programming and infrastructure projects. Currently, as a key member of the Machine Learning (ML) Compiler Team for Augmented Reality (AR) and Virtual Reality (VR), Riley is at the...
Riley Dulin is a seasoned software engineer at Meta, where he has honed his expertise in systems-level programming and infrastructure projects. Currently, as a key member of the Machine Learning (ML) Compiler Team for Augmented Reality (AR) and Virtual Reality (VR), Riley is at the forefront of developing innovative solutions that enable efficient execution of PyTorch ML models on power-constrained devices. His work focuses on optimizing computer vision tasks, a critical component in enhancing user experiences in AR/VR environments. By leveraging advanced techniques such as 3D convolutions and utilizing torch.fx as an intermediate representation (IR), Riley is instrumental in pushing the boundaries of what is possible in mobile and edge computing.
Prior to his current role, Riley contributed significantly to the Hermes team from 2017 to 2021, where he worked on an open-source JavaScript engine designed to improve performance in mobile applications. His experience in this area has equipped him with a robust understanding of web development and the intricacies of JavaScript, further enhancing his versatility as a software engineer.
Riley's technical skill set is diverse, encompassing compilers, machine learning, and various programming languages including Go and C++. His proficiency in databases such as MySQL and MongoDB complements his systems-level programming capabilities, allowing him to design and implement comprehensive solutions that address complex challenges. With a passion for infrastructure projects and a commitment to advancing technology, Riley Dulin continues to make significant contributions to the evolving landscape of software engineering at Meta.
