Cody Coleman, co-founder and CEO of Coactive AI, explains how they leverage modern data, systems, and machine learning techniques for their multimodal asset platform and visual search tools. They discuss techniques like active learning and core set selection, and how they drive efficiency throughout the machine learning lifecycle. Cody also shares how Coactive uses multimodal embeddings for visual search and the infrastructure optimizations they've implemented to scale their systems. They conclude with advice for entrepreneurs and engineers in generative AI technologies.
Read more
AI Summary
Highlights
AI Chapters
Episode notes
auto_awesome
Podcast summary created with Snipd AI
Quick takeaways
Coactive AI leverages embeddings to simplify and accelerate machine learning development, enabling a more agile and iterative approach.
The partnership with Amazon Web Services provides support in scalability, security, and go-to-market strategies, contributing to the success of Coactive AI.
Deep dives
Coactive AI: Simplifying and Scaling Multimodal AI
Coactive AI is a multimodal asset platform that simplifies and scales the process of searching and analyzing content. It enables a shift from a waterfall approach to a more agile approach in machine learning development by leveraging embeddings as a cash for computation. It addresses the challenges of scaling with the massive amount of unstructured data by providing a system that can handle data oceans instead of just data lakes. Security, model agnosticism, and cloud agnosticism are key considerations for enterprises, and Coactive AI offers solutions in all these areas. The platform enables companies to future-proof themselves by being adaptable to evolving AI models and being able to read data from various sources. The partnership with Amazon Web Services has been instrumental in the success of Coactive AI, providing support in several areas including scalability, security, and go-to-market strategies.
Using Active Learning and Dynamic Tagging to Automate Labeling
Coactive AI uses active learning and dynamic tagging techniques to automate the labeling process. Active learning is leveraged to find the most informative data points for annotation, resulting in significant time and cost savings. By iteratively selecting and labeling relevant examples, the system fine-tunes models and generates consistent metadata over the entire content catalog. This automates tasks that were previously done manually, such as reviewing images for compliance with community guidelines. The platform has demonstrated an 85% reduction in manual labeling efforts for a large entertainment and gaming platform. Additionally, the use of multimodal embeddings enables faster and more accurate retrieval and analysis of content.
Harnessing Embeddings as a Driver for Simplicity and Agile ML
Coactive AI utilizes embeddings as a core component for simplifying and accelerating machine learning development. By leveraging pre-trained models and embedding vectors, the platform decouples the slow and expensive deep learning process from downstream tasks. This enables a more agile and iterative approach to developing ML models. By shifting from a waterfall approach to an embedding-based approach, companies can quickly prototype and fine-tune models without having to go through multiple layers of training and annotation. Embeddings also facilitate fast and scalable processing of large-scale unstructured data, enabling businesses to unlock valuable insights from their content. By embracing embeddings, Coactive AI paves the way for future advancements in AI models and methods.
The Importance of Scale, Security, and Future-Proofing in AI
The podcast highlights the importance of scale, security, and future-proofing in the context of AI. With the exponential growth of unstructured data, companies need systems that can handle large-scale content and leverage AI advancements without sacrificing security. Coactive AI addresses these needs by providing a platform that can handle data oceans and handle different AI models without being tied to a specific vendor or cloud provider. This approach allows businesses to adapt to evolving AI technologies and leverage the best models for their specific use cases. Furthermore, ensuring data security and compliance is crucial in the era of AI, and Coactive AI prioritizes data governance and privacy to meet enterprise requirements. By embracing scalability, security, and future-proofing, companies can harness the full potential of AI for their business success.
Today we’re joined by Cody Coleman, co-founder and CEO of Coactive AI. In our conversation with Cody, we discuss how Coactive has leveraged modern data, systems, and machine learning techniques to deliver its multimodal asset platform and visual search tools. Cody shares his expertise in the area of data-centric AI, and we dig into techniques like active learning and core set selection, and how they can drive greater efficiency throughout the machine learning lifecycle. We explore the various ways Coactive uses multimodal embeddings to enable their core visual search experience, and we cover the infrastructure optimizations they’ve implemented in order to scale their systems. We conclude with Cody’s advice for entrepreneurs and engineers building companies around generative AI technologies.
The complete show notes for this episode can be found at twimlai.com/go/660.
Get the Snipd podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered podcast player
Listen to all your favourite podcasts with AI-powered features
Discover highlights
Listen to the best highlights from the podcasts you love and dive into the full episode
Save any moment
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Share & Export
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
AI-powered podcast player
Listen to all your favourite podcasts with AI-powered features
Discover highlights
Listen to the best highlights from the podcasts you love and dive into the full episode