Energy-Efficient AI: Building Sustainable Data Pipelines for the Future
Artificial Intelligence is driving innovation across industries, but its growing energy demands pose critical challenges around cost, scalability, and sustainability. In this talk, Bhanu will share practical strategies for designing energy-efficient AI systems, focusing on: dynamic batching & KV caching for reducing inference overhead, sparse neural networks & structured pruning for lightweight models, carbon-aware scheduling to align compute with renewable energy, federated learning & edge deployments to reduce data transfer energy, and a sustainability maturity model for organizations adopting Green AI. The session combines real-world case studies, industry frameworks, and research-driven practices to show how you can optimize AI pipelines for both performance and sustainability. Attendees will leave with actionable approaches to measure, monitor, and minimize the carbon footprint of AI workloads, making sustainability a core design principle rather than an afterthought.
Bhanu Prakash Reddy Rella is a Senior Data Engineer at Meta, specializing in large-scale data infrastructure and sustainable AI systems. With over a decade of experience spanning Meta and Walmart, he has designed and optimized high-volume data pipelines, storage frameworks, and processing systems that reduce energy consumption and computational waste. He is the co-author of Energy-Efficient Computing for Modern AI Applications (2025), which provides practical strategies such as federated learning, structured pruning, and carbon-aware scheduling for building sustainable intelligence. Currently pursuing a Doctor of Business Administration at Golden Gate University, his research focuses on embedding sustainability principles into AI testing, evaluation, and deployment frameworks. Rella has chaired IEEE sessions, judged international hackathons, and delivered keynotes on Green AI. Through his leadership of The Green AI Initiative, he advocates for aligning technological innovation with ecological responsibility.