Kai Malcolm

About Kai Malcolm

I am a fourth-year PhD Candidate at Rice University, building high-bandwidth, intuitive neuromotor interfaces. My research is centered on developing the algorithms and engineering pipelines required to make biosignal-based interfaces (e.g, EMG muscle-controlled interfaces) reliable, private, and personalized for real-world medical and wearable applications.My current project focuses on developing one-shot personalized gesture classifiers for users with upper-body motor impairments. To achieve this, I design and deploy complex ML pipelines, incorporating CNN-LSTM architectures with Mixture-of-Experts (MoE) for soft joint clustering and meta-learning to ensure rapid generalization across novel users.I specialize in the ML engineering side of neural interfaces: building robust data augmentation pipelines, fine-tuning models, and optimizing few-shot learning protocols to reduce the calibration burden on the end-user. Key Expertise: Time Series Physiological Modeling, Personalization, Meta-Learning, Deep Learning (PyTorch), and Federated Learning.

Experience

1. Machine Learning Intern

   Tacta Systems, Inc

   Jun 2026 — Present · Palo Alto, CA, US

Education

• Vanderbilt University

  Bachelor of Engineering - BE, Mechanical Engineering, Minors in Scientific Computing, Electrical Engineering

  2019 — 2022

• Rice University

  Master of Science - MS, Electrical and Computer Engineering

• Rice University

  Doctor of Philosophy - PhD, Electrical and Computer Engineering

• Shawnee High School

  High School Diploma with Honors

  2014 — 2018

• Case Western Reserve University

  Bachelor of Arts - BA, Chemical Engineering

  2018 — 2019