My guest today is Chu-Cheng Hsieh, Chief Data Officer (CDO) at Etsy -- a global online marketplace, where people come together to make, sell, buy, and collect unique and creative items. Chu-Cheng leads the global data organization responsible for data science strategy, AI innovation, machine learning & data infrastructure. Prior to Etsy, Chu-Cheng led various data roles at Amazon, Intuit, Rakuten, and eBay. Chu-Cheng is a Ph.D. in computer science, with published papers in key AI/ML conferences.
In this episode, I chat with Chu-Cheng on a wide range of topics: Evolution of recommendation algorithms and dealing with the cold start problem w.r.t. user preferences; Experience related to data labeling including version control of labeling rules/taxonomies; Model building taking into account seasonality, drift, etc.; Handling data accuracy/noise and defining success criteria; Battlescars in model evolution; Team structure, hiring, and diversity; Aspects to make data and AI self-service for technical and business users.
My guest today is Amit Aggarwal, the CTO of The YES -- a retail app and platform that focuses on women's fashion and clothing. The YES optimizes product personalization using AI/ML. In this fascinating talk, Amit shares insights on a broad range of topics from technology building blocks, measuring success criteria of AI, re-training models to keep up with changing fashion trends, team hiring, and key lessons learnt in the journey.
In this episode, my guest is Daniel Mccaffery. Daniel is a technology thought leader driving Data and Analytics at Climate Corporation (a division of Bayer).
Daniel shares his insights on using ML to provide personalized recommendations for helping farmers grow crops with higher yield, profitable and sustainability. This involves deciding the right seed, right crop protection, density levels across different parts of the farm, etc. This is a fascinating example of AI and physical sciences coming together to build an innovative product offering. Daniel and I had a blast covering several topics: the building of models, model deployment and re-training, explainability for farmers to understand the recommendations, managing bias, experimentation A/B testing, monitoring drifts, data labeling, and perspectives on key bottlenecks in going from idea to ROI.
