Carnegie Mellon University’s Robotics Institute is seeking a Research Assistant to investigate new algorithms and models for analyzing and understanding human behavior from video, focusing on the design of efficient perceptual algorithms for behavioral cue extraction and novel approaches for the modeling people interaction, with application to medical research and affective computing.
Requirements
- Strong programming skills, Python, Pytorch, Large Vision Models, Multimodal Foundation Models, Transfer Learning.
- Master or bachelor degree in computer science, applied mathematics, electrical and/or computer engineering with a focus on computer vision/machine learning.
- Three years’ experience in a research environment with at least one year in the designated specialty of research.
Responsibilities
- Write code and develop novel theoretical and practical state of the art artificial intelligence/machine learning algorithms that are focused on human behavior modeling related to video classification using deep learning networks for end-users.
- Work with other team members to develop and maintain software for maximum efficiency and usability.
- Designs, develops, tests and maintains applications or system programs.
- Gathers and provides input on functional specifications and requirements
- Perform approved research project tasks according to defined protocols and procedures including appropriate collection, recording, storage, and analysis of data, and reporting of results.
- Follow appropriate research policies and procedures and maintains scientific integrity including accurate data storage.
- Optimize protocols and improve methods currently employed.
Other
- Master or bachelor degree in computer science, applied mathematics, electrical and/or computer engineering with a focus on computer vision/machine learning.
- Strong interpersonal, verbal, and written communication skills.
- Strong organization and planning skills, careful attention to detail with strong follow-through, able to prioritize and organize tasks effectively to accomplish objectives in a timely matter.
- Enthusiastic about collaborating with partners from multiple disciplines and institutions.
- OSHA Hazardous Materials Safety Training