Grace Adamson: Building End to End Gen AI Powered Analytics: From Data to Production - Codemotion 25
Operationalizing machine learning isn’t just about building a model—it’s about creating a reliable, scalable pipeline from raw data to real-time inference. In this talk, we’ll walk through an end-to-end ML workflow in Python designed for developers, data scientists, and ML engineers who want to move fast and build production-ready systems without reinventing the wheel. You’ll learn how to: - Prepare data and engineer features using consistent, reusable patterns to avoid duplication and drift across training and inference. - Train and tune models with popular open-source libraries like scikit-learn, XGBoost, and LightGBM, on CPU or GPU. - Package and deploy models for real-time or batch inference with minimal ops overhead. - Track experiments, monitor performance, and debug issues with built-in observability, lineage tracking, and model explainability. We’ll show how all of this can be done within a unified workflow using Python, with the help of containerized runtimes and built-in versioning, orchestration, and deployment tools—so you can focus on solving problems, not managing infrastructure. This is a practical, hands-on session for developers who want to go from notebook to production without duct tape. By the end, you’ll walk away with a practical framework for building resilient ML systems that scale. Puedes seguir a Codemotion en sus redes sociales: @Codemotionworld // https://www.linkedin.com/company/codemotion/ // https://twitter.com/CodemoMadrid // https://www.instagram.com/codemotion_esp/ ------- Síguenos en nuestras redes sociales Web: https://sirviendocodigo.com/ LinkedIn: https://www.linkedin.com/company/sirviendo-codigo/ X: https://twitter.com/sirviendocodigo Instagram: https://www.instagram.com/sirviendo.codigo/ TikTok: https://www.tiktok.com/@sirviendo.codigo #software #softwarearchitecture #code #sirviendocodigo
