Sign In
Register

Partnership opportunities

Secure your seat

Call to action
Your text goes here. Insert your content, thoughts, or information in this space.
Button

Back to speakers

Dr. Dmitry
Kazhdan, PhD
CTO & Co-Founder
Tenyks
Dr. Kazhdan is the Co-Founder & CTO at Tenyks, a University of Cambridge spinout specializing in Visual Intelligence. As a Co-Founder, Dr. Kazhdan has played a pivotal role in shaping the company's technological foundations. In the past several years, Tenyks has achieved significant milestones, including graduating from the Y Combinator accelerator, and being named Cambridge's Company of the Year in 2022. Currently, Dr. Kazhdan and the Tenyks team are busy developing a platform for Visual Intelligence, capable of processing vast amounts of video data using state-of-the-art Foundation Models quickly and cheaply, enabling users to query and extract structure from visual data using natural language. Dr. Kazhdan's earlier work, undertaken during his PhD studies at Cambridge, focused on the safety and explainability of Deep Learning systems across modalities. His research, which includes several notable papers on Concept-based Explanations for visual models, has laid the groundwork for innovation driving some of Tenyks' present advancements.
Button
07 November 2024 12:00 - 12:30
"Can't I just ask ChatGPT?" Production pipelines shaping the visual intelligence renaissance
We are entering a new era of Visual Intelligence, where the need for training custom models from scratch is rapidly diminishing. Today, many vision tasks can be accomplished simply by leveraging pre-existing models (such as Foundation Models, MLLMs, or Open Vocabulary models), fine-tuned or taken directly off-the-shelf. Crucially, such models are becoming increasingly more powerful and entirely sufficient for a wide range of applications, including Retail Analytics, Surveillance, Media & Entertainment, Logistics, and many more. While such models have achieved remarkable progress, several challenges remain in their integration into production pipelines, including: high operational costs at scale, performance variability across tasks, security concerns, and integration challenges. As a result, there is currently a big gap between the capabilities of these models and fully developed, production-ready vision systems. In this talk, we will discuss how to overcome these challenges and implement robust, scalable computer vision pipelines powered by the latest AI model advancements.