Embedded AI - Research Engineer

The Audio team within RL Research is looking for a Researcher Engineer with expertise in efficient multimodal machine learning and embedded systems to join our team. You will be building technologies that improve the listener's hearing under challenging listening conditions using wearable computing. We are looking for expertise in building efficient representation models using audio, visual and speech signals; bridge and integrate AI models with conversational context and domain understanding. You will operate at the intersection of large-scale machine learning systems design and development, and egocentric audio-visual learning and computer vision, and partnering with experts in systems processing; and hardware/software co-design.

Embedded AI - Research Engineer Responsibilities:

• Work with AI researchers and audio/acoustics domain experts on designing and building novel low-compute, low-power ML and CV for egocentric audio-visual learning.
• Optimize, Implement and benchmark ML model on ML accelerators.
• Profile AI model run-time inference across various metrics, e.g. compute, memory, power and latency.
• Support quick prototyping, proof of concept, or proof-of-experience and demonstrations via the real-time integration of ML and CV models into research & development platforms for wearables.
• Develop and maintain technical documentation for AI model implementation and optimization.

• Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience
• Familiarity with SoC Architecture: System-level understanding of data flows and memory management, including simulation of hardware components and functions such as NPU, GPU, CPUs and DSPs.
• 2+ years of experience in developing, implementing, and optimizing real-time AI models on edge devices.
• Proficiency in leveraging popular AI engines such to optimize AI-driven applications for camera and SoC platforms.
• In-depth knowledge of the architecture and programming models of popular AI engines, enabling effective utilization of their features and resources to achieve high-performance and power-efficient AI processing.
• Familiarity with relevant tools such as ExecuTorch, TensorFlow Lite, or similar frameworks.
• Demonstrated experience in programming skills in C++, Python, or other relevant languages.
• Experience with cross-group and cross-discipline collaboration.

• Master's Degree in Computer Sciences, Computer Engineering, Deep Learning, Artificial Intelligence, Machine Learning, Robotics, Computer Vision, Computational Neuroscience, Signal Processing, Speech and Language technologies, or a related field, or equivalent practical experience.
• 4+ years of experience in developing, implementing, and optimizing real-time AI models on edge devices.
• 3+ years of experience working on efficient machine learning or computer vision algorithms.
• Experience with end-to-end real-time ML pipelines, large-scale ML benchmarking, real-time statistical modeling including heuristics driven computer vision methods.
• Experience working on evaluation and benchmarking for audio-visual ML models, or related generative AI models.
• Experience working with datasets on preprocessing methods, dataloaders, data tooling and related software engineering platforms.
• Experience with large-scale or distributed cluster computing for training, development and offline inference of machine learning models.
• Experience working and communicating cross functionally in a team environment.
• Experience solving complex problems and comparing alternative solutions, tradeoffs, and broad points of view to determine a path forward.
• Proven track record of achieving significant results as demonstrated by project launch, research grants, fellowships, patents, as well as publications at leading workshops, journals or conferences.