About The Position
Develop real-time deep learning and signal processing solutions for sensing and imaging systems, including video restoration, as part of an interdisciplinary team working on technologies for future Apple products. The role covers algorithm and model development using real sensor data, simulated data, and public datasets; dataset preparation and analysis; benchmarking; and optimization for low latency and resource-efficient on-device performance. The engineer combines classical signal processing with learned models, investigates architectures for real-time inference, and explores ways to mitigate limitations and artifacts of sensing and imaging systems. The role also involves collaboration with ML researchers and software, hardware, sensor, and architecture engineers to turn research into practical systems, as well as technical guidance and mentorship for junior engineers.
Requirements
- Strong foundation in signal processing, image processing, computer vision, machine learning, and deep learning.
- Hands-on experience with sensing technologies such as camera/image sensors, depth, radar, LiDAR, IMU, or other sensor modalities, including understanding sensor noise models and characteristics.
- Demonstrated experience designing and deploying real-time deep learning and signal processing solutions using large-scale datasets.
- Proficiency in Python and deep learning frameworks such as PyTorch.
- Experience with model evaluation, debugging, experimental analysis, and failure analysis.
- PhD or Master's degree in Electrical Engineering, Computer Science, or a related research field.
- Proficient written and verbal communication skills.
- Ability to work autonomously and collaboratively.
- 8+ years of relevant experience.
- Experience deploying models on GPU, dedicated hardware, or neural engines, including quantization, pruning, and other model-compression techniques.
Nice To Haves
- Experience with C++ or other performance-oriented languages.
- Experience with multi-sensor fusion, sensor calibration, or temporal/streaming signal processing.
- Experience with generative priors such as diffusion or flow matching.
- Background in signal processing, physics, optics, or lasers.
Responsibilities
- Design, implement, and optimize signal processing and machine learning models for sensing and imaging tasks, including video restoration.
- Develop and validate models using real sensor data, simulated data, and public datasets.
- Prepare, filter, process, and analyze large-scale datasets for model training and evaluation.
- Analyze algorithm results and benchmark them against leading methods and theoretical bounds.
- Optimize models for real-time performance, low latency, and resource efficiency on-device.
- Develop algorithms combining classical signal processing with learned models to extract robust information from sensor data.
- Investigate efficient architecture patterns suitable for real-time, on-device inference.
- Explore algorithmic approaches to mitigate artifacts and limitations of new sensing and imaging systems.
- Stay current with advances in signal processing, computer vision, and machine learning.
- Collaborate with ML researchers and software, hardware, sensor, and architecture engineers to translate research into robust models and practical systems.
- Provide technical guidance and mentorship to junior engineers and help shape the team's technical direction.
- Own end-to-end development of real-time algorithms from concept and prototyping through validation and deployment.
- Define evaluation methodology, metrics, and failure-analysis processes for sensing and imaging algorithms.
- Drive trade-off decisions among accuracy, latency, power, and memory across algorithm, hardware, and system constraints.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
What This Job Offers
Job Type
Full-time
Career Level
Senior
