Revolutionizing Retail with Vision-Language Models: AI-Powered Shopping

Discover how cutting-edge vision-language models are revolutionizing the digital shopping experience in this insightful session. Learn how fine-tuned versions of OpenAI’s CLIP and FLAVA models powered two game-changing applications: an AI-driven Style Filters platform that boosted customer engagement and an Automated Attribute Correction system that improved product discovery accuracy. Dive into the technical challenges of fine-tuning these models for retail, explore advanced prompt engineering strategies, and uncover scalable solutions that serve millions of customers. Gain practical insights to elevate AI-driven retail innovation at scale!

Key Takeaways:

• Learn how to fine-tune vision-language models for retail-specific applications.
• Discover techniques for implementing automated product attribution systems to enhance efficiency.
• Understand strategies for scaling AI solutions in large retail environments.
• Discover methods to measure and maximize the business impact of AI implementations.

Shubham is an accomplished Data Scientist with expertise in developing large-scale AI solutions across retail, healthcare, and finance sectors. Currently, at Nordstrom Inc, he leads the development of innovative AI applications using vision-language models. His work focuses on fine-tuning large language models, computer vision, and generative AI, emphasizing creating scalable, business-impacting solutions. Before Nordstrom, he held ML engineering-based roles at Amazon and HealthTap. Shubham is passionate about democratizing AI development by establishing robust standards and mentoring next-generation data scientists. You can reach him on LinkedIn.