Career
Present: Co-Founder & CEO at Claryo.
Past: Co-Founder & CSO at Robust AI.
Past: Senior Technical Manager at SRI International.
Education
Ph.D., Electrical and Computer Engineering, OSU.
Dissertation: Generative hierarchical graphical models for human activity recognition in large-scale videos.
M.Sc., Electrical and Computer Engineering, OSU.
Thesis: Monocular extraction of 2.1D sketch and multi-view 3D reconstruction.
B.Sc., Electrical and Computer Engineering, OSU.
Project: Tele-Robotic Arm Controller.
Aviation
License: Private Pilot.
Ratings: Instrument and high-performance.
Claryo is revolutionizing warehouse operations using Spatial Generative AI by integrating data from various sources and creating accurate virtual representations. Claryo helps warehouse operators overcome operational bottlenecks, improve efficiency, and supports the seamless integration of robotic systems.
Carter is a state-of-the-art robot that I helped develop at Robust AI, designed to push the boundaries of AI and robotics with a strong emphasis on human-robot interaction. Equipped with advanced artificial intelligence, Carter excels at performing complex tasks in dynamic environments with remarkable efficiency and adaptability. My primary focus was on defining the roadmap and initial implementation of the perceptual enrollment processes for Carter ensuring seamless deployment in target facilities.
This project explores the use of AI to enhance the design process of building facades. Leveraging generative AI techniques, the project aims to automate and optimize the design of building exteriors, improving aesthetics, functionality, and energy efficiency. By integrating AI into the design workflow, the project seeks to streamline the creative process, allowing for more innovative and practical facade solutions in architecture.
This program seeks to create AI systems that can explain their decisions and actions to human users. By enhancing transparency and trust in machine learning models, XAI aims to improve the collaboration between humans and AI systems, making AI's decision-making processes more understandable, predictable, and controllable for users. The focus of my research was to enhance the explainability of generative AI models and understand the effect of explainability on the users' trust levels.
This program aims to develop new ways for humans and machines to interact using natural language and dialogue while reasoning over complex domains. The goal is to create systems where computers can understand and respond to human communication more effectively, moving beyond simple command-based interfaces to more conversational and collaborative interactions. The focus of my research was to build AI models for collaborative composition for visual storytelling, by analyzing movies, videos, and scripts to generate short animations from text.
This program aims to develop neural-inspired computing architectures that mimic the brain’s learning and pattern recognition abilities. The goal is to create real-time, adaptive systems capable of processing complex data streams with minimal training. By leveraging low-power, brain-inspired algorithms, the program seeks to address challenges in recognizing and acting on complex structures in data, enabling breakthroughs in applications such as anomaly detection and prediction. The focus of my research was to build a unified machine learning model that is capable of reasoning over multimodal, time-series, sensory data. In addition, the model should be memory and energy efficient for deployment on FPGAs.
This program focuses on developing tools to improve human interaction in high-stakes environments. The goal is to create systems that enhance communication, trust, and rapport-building in situations such as military engagements or negotiations, where social dynamics are critical. SSIM seeks to provide strategies for more effective interpersonal interactions to improve mission outcomes. The focus of my research was creating interaction datasets and analyzing human social interaction using multimodal machine learning models over complex multisensor data.
This program aims to advance the mathematical foundations of artificial intelligence and machine learning. It focuses on integrating sensing, perception, and action for autonomous systems to interpret and adapt to complex environments. The program’s goal is to develop systems that can intelligently respond to sensory input by executing actions based on sophisticated reasoning in real time. The focus of my research was on building computer vision models to recognize human activity in large scale videos.