Ads Ranking Evolution at Pinterest // Aayush Mudgal // #211

Aayush Mudgal is a Senior Machine Learning Engineer at Pinterest, currently leading the efforts around Privacy-Aware Conversion Modeling.Large Language Models have taken the world by storm. But what are the real use cases? What are the challenges in productionizing them? In this event, you will hear from practitioners about how they are dealing with things such as cost optimization, latency requirements, trust of output, and debugging. You will also get the opportunity to join workshops that will teach you how to set up your use cases and skip over all the headaches.Join the AI in Production Conference on February 15 and 22 here: https://home.mlops.community/home/events/ai-in-production-2024-02-15_________________________________________________________MLOps podcast #211 with Aayush Mudgal, Senior Machine Learning Engineer at Pinterest, Ads Ranking Evolution at Pinterest.// AbstractListen to the lessons from the journey of scaling ad ranking at Pinterest using innovative machine learning algorithms and innovation in the ML platform. Learn how they transitioned from traditional logistic regressions to deep learning-based transformer models, incorporating sequential signals, multi-task learning, and transfer learning. Discover the hurdles Pinterest overcame and the insights they gained in this talk, as Aayush shares the transformation of ad ranking at Pinterest and the lessons learned along the way. Discover how ML Platform evolution is crucial for algorithmic advancements.// BioAayush Mudgal is a Senior Machine Learning Engineer at Pinterest, currently leading the efforts around Privacy-Aware Conversion Modeling. He has a successful track record of starting and executing 0 to 1 projects, including conversion optimization, video ads ranking, landing page optimization, and evolving the ads ranking from GBDT to DNN stack. His expertise is in large-scale recommendation systems, personalization, and ad marketplaces. Before entering the industry, Aayush conducted research on intelligent tutoring systems, developing data-driven feedback to aid students in learning computer programming. He holds a Master's in Computer Science from Columbia University and a Bachelor of Technology in Computer Science from the Indian Institute of Technology Kanpur.// MLOps Jobs board jobs.mlops.community// MLOps Swag/Merchhttps://mlops-community.myshopify.com/// Related Linkshttps://www.youtube.com/watch?v=MZVIxtsGzBg https://www.youtube.com/watch?v=ffpPUr8Hg6U --------------- ✌️Connect With Us ✌️ -------------Join our Slack community: https://go.mlops.community/slackFollow us on Twitter: @mlopscommunitySign up for the next meetup: https://go.mlops.community/registerCatch all episodes, blogs, newsletters, and more: https://mlops.community/Connect with Demetrios on LinkedIn: https://www.linkedin.com/in/dpbrinkm/Connect with Aayush on LinkedIn: https://www.linkedin.com/in/aayushmudgal/Timestamps:[00:00] Join the AI in Production Conference![02:10] Aayush preferred coffee[02:31] Takeaways[03:41] Evolving decision-making based on evolution and ROI[09:53] Companies have varying approaches to building[13:54] Continuous innovation and adaptation[17:13] Transform, train, and analyze data for effective predictions[21:01] Shift in traditional systems, monitoring, and visibility[22:36] Monitoring all pipelines, models, features, and predictions. Alerting[24:56] Maintain simplicity and optimize pipelines for scaling[28:40] Check if systems are ready for change[34:13] Commitment, tooling, and understanding are crucial for migration[36:30] Concerns about technology support and migration strategy[41:44] Difficulty removing hybrid systems, but speed benefit.[45:39] Recommendation models learn user-content interactions, transformer as a feature interaction layer[49:07] Optimize model complexity, control sequence length, and reduce costs[51:01] Pinterest uses Pytorch for training and complex serving[52:14] Wrap up