Research Scientist Intern, Wearable Sensing & Human Motion Analysis

About the role#

This internship focuses on building learned models for motion analysis in swing sports. You will replace heuristic, threshold-based methods with models that improve accuracy, handle different players and equipment, and provide interpretable feedback. You will work within the Surreal Vision group to build egocentric machine perception functionalities for future AR and VR devices.

What you'll do#

• Operate egocentric data collections using an in-lab simulator.
• Instrument participants with Aria Gen 2 and Neural Band wearables.
• Run capture sessions and use simulator ground-truth as a signal for model validation.
• Design and train models over multimodal egocentric time-series data.
• Solve problems including temporal swing phase segmentation and kinematic estimation from head-mounted and wrist-worn viewpoints.
• Build an evaluation harness to benchmark models against existing baselines.
• Define capture and labeling protocols while managing small, noisy, and personalized datasets.
• Integrate your models into prototypes for end-to-end demonstrations.
• Write up your results for potential publication or internal technical reports.

What you'll need#

• Current enrollment in or pursuit of a PhD in Computer Science, Electrical Engineering, Robotics, Biomechanics, or a related field.
• Strong Python skills and hands-on experience training models in PyTorch.
• Working knowledge of sequence and time-series modeling, such as temporal CNNs, transformers, state-space models, or HMM/CTC-style segmentation.
• Experience working with real sensor data rather than only curated benchmarks.
• Comfort running hardware experiments with human participants in a lab setting.
• Ability to obtain and maintain work authorization in the United States.
• Experience designing and executing data collections with wearable or motion-capture hardware.
• Research experience with IMU-based human activity recognition, pose estimation, or wearable biosignals.
• Experience with surface EMG, including filtering, rectification, and handling electrode-shift variability.
• Familiarity with egocentric perception, multimodal sensor fusion, or sports biomechanics.
• Experience with self-supervised or few-shot learning for personalization.
• First-author publications in relevant venues like CVPR, ICCV, ECCV, NeurIPS, or ICLR.
• Personal experience playing swing sports is helpful for labeling and sanity-checking feedback.
• Intent to return to your degree program after the internship.

Location & details#

• Location: Redmond, Washington.
• Work modality: On-site.
• Duration: 12 to 24 weeks.
• Employment type: Full-time internship.
• Compensation: $7,650 to $12,667 per month.