Piaggio Fast Forward conducts observational and laboratory research on the ways that people and things move in and around pedestrian environments. The resulting data helps establish performance specifications for vehicles designed to operate in these same environments while interacting with human operators and bystanders. Our data is granular and objective, using high-rate motion capture to break down body language, gait, posture, social proximity, and many other aspects of non-verbal communication in the public sphere.

This position calls for someone who seeks to contribute to and learn from a company that utilizes cutting-edge data collection, maintenance, and analysis methods for the purposes of creating supervised and unsupervised models of human-robot behavior, interaction, and social etiquette. We’re looking for someone who is passionate about data clarity and the reduction of bias in new and groundbreaking domains, including generative, causal, and multimodal AI models. The ideal candidate is innovative, highly organized, and possesses strong analytical skills, able not only to recognize and define patterns in datasets, but also to help us maintain and update the cleanest and most comprehensive dataset possible.  

This role is an integral part of the R+D team. Our technology identifies and tracks humans while building maps and recording paths on edge devices using a fusion of image, radar, and Lidar sensing. In addition to the edge computing data from our mobile robots, we have a proprietary database of submillimeter detailed, organized, and tagged human movement in the built environment, as well as a database of unstructured video from our testing robots in the real world. Due to this abundance of data, we are seeking candidates with experience in the use of generative AI, VLMs, or other computer vision ML technologies for the semantic understanding of scenes that can help us leverage this data.

The research conducted and the reports generated by the Smart Behaviors AI Co-op, while valuable to the company as a whole, will ultimately serve to enhance the day-to-day experience of our individual consumer and commercial customers.

 Responsibilities

• Work with diverse datasets, including motion capture data and video recordings
• Research and implement various AI methods in both the capture and classification of large data sets, including Causal AI, Generative AI, and Multimodal AI
• Help maintain highly organized databases with robust classifications
• Document results in detailed data visualizations 
• Provide statistical evaluation of various results and analysis methods to verify accuracy

Requirements

• Pursuing a Bachelor’s or Master’s degree in Data Engineering, Data Science, or something similar
• Knowledge of or experience in research and development methodologies
• Knowledge of or experience in creating and utilizing generative AI and VLMs for the semantic understanding of unknown scenes
• Knowledge of or experience in data aggregation, processing, and automation of workflows
• Experience in Python, MATLAB, and/or R for analysis, statistical evaluation, and visualization (Plotly, Matplotlib, Pandas, Numpy, etc)
• Highly organized and have impeccable attention to detail
• Diligent in sorting, categorizing and labeling large volumes of data
• Preferred:
• Experience with motion capture software and hardware
• High level of curiosity and drive for innovation