This episode discusses how the team develops a Generative AI solution to create engaging email subject lines by combining a reward model and prompt engineering over ChatGPT APIs.

Based on the tech blog shared by Machine Learning Engineers from Nextdoor, the blog link is: https://engblog.nextdoor.com/let-ai-entertain-you-increasing-user-engagement-with-generative-ai-and-rejection-sampling-50a402264f56

This episode discusses what is customer segmentation and how to use the k-means clustering algorithm over Recency, Frequency, and Monetary (RFM) segmentation features to generate them algorithmically.

Based on the tech blog shared by a Data Scientist from Microsoft, the blog link is: https://medium.com/data-science-at-microsoft/introduction-to-clustering-based-customer-segmentation-2fac61e80100

This episode discusses how contrastive learning (one self-supervised machine learning algorithm) is used to enable in-video search capability, and help facilitate creating promotional videos.

Based on the tech blog shared by Machine Learning Engineers from Netflix, the blog link is: https://netflixtechblog.com/building-in-video-search-936766f0017c

This episode discusses what is geo experimentation, and how synthetic control methods can be used to help make it more efficient, along with some tips and good practices shared by the blogpost author.

Based on the tech blog shared by Data Scientists from Mercado Libre, the blog link is: ⁠⁠⁠⁠⁠⁠⁠https://medium.com/mercadolibre-tech/harnessing-the-power-of-geo-experimentation-how-mercado-libre-measures-the-effectiveness-of-its-f68b38857c4b

This episode discusses the importance of faster experimentation velocity and how one powerful methodology, CUPED (Controlled-experiment Using Pre-Experiment Data), can be used to achieve the goal.

Based on the tech blog shared by Data Scientists from Statsig, the blog link is: ⁠⁠⁠⁠https://www.statsig.com/blog/cuped

This episode discusses how math and data science approaches make geospatial search easy.

Based on the tech blog shared by Data Scientists from Walmart, the blog link is: ⁠⁠https://medium.com/walmartglobaltech/geospatial-search-made-easy-52c0f213ea93

This episode discusses how the New York Times cooking team made personalized recipe recommendations.

Based on the tech blog shared by Data Scientists from The New York Times, the blog link is: ⁠https://medium.com/@timesopen/how-the-new-york-times-cooking-team-makes-personalized-recipe-recommendations-669c26aa4825

This episode discusses how synthetic search data is leveraged to enable Expedia's flight price forecasting product.

Based on the tech blog shared by Data Scientists from Expedia, the blog link is: https://medium.com/expedia-group-tech/using-synthetic-search-data-for-flights-price-forecasting-4cf3277afdaf

In this episode, we will discuss Lable Noise problems and solutions as shared by Data Scientists from Walmart

Their tech blog link: https://medium.com/walmartglobaltech/label-noise-problems-and-solutions-5e38e03dd74

In this episode, we will discuss how Instacart provides incrementality measures in its advertising business, and some related data science concepts, such as randomized control trials.

Their tech blog link: https://tech.instacart.com/how-instacart-measures-the-true-value-of-advertising-the-methodology-of-ad-incrementality-aac6c58e627c