Nilesh leads engineering for Inferless, a stateful serverless platform built from scratch to help developers deploy custom and open source models with low cold starts and efficient autoscaling. He has helped hundreds of companies optimize their cloud infrastructure; developed a machine learning-based ads automation platform while at Amazon; and has published a patent in using containerization of applications and several others on code analysis.
Nilesh’s expertise lies in machine learning infrastructure, Kubernetes, engineering, AI infrastructure, AI model deployment and observability and AI agent deployment. He holds a Bachelor of Technology in Computer Software Engineering from Delhi College of Engineering.
On this episode, Dr. Darren engages in a stimulating conversation with Nilesh Agarwal, co-founder and CTO of InferLess. Nilesh explores the evolution of AI and the crucial role of data management in the current landscape. He highlights the challenges organizations face in terms of data security, efficiency, and the need for innovative data architectures. The discussion also delves into the significance of edge computing, the potential of hybrid AI models, and the emergence of specialized hardware to meet the evolving demands of AI applications. Nilesh emphasizes the importance of integrating AI into data pipelines to improve data access and security, while addressing the complexities of managing multiple models and ensuring the efficient use of compute resources. ## Takeaways * AI has shifted the focus from compute to data management. * Data efficiency is crucial for effective model training. * Organizations are increasingly concerned about data security. * Data warehouses are often inadequate for modern data needs. * New architectures, such as vector databases, are emerging. * AI can enhance data access through natural language queries. * Hybrid models will dominate the future of AI.. * Edge computing is essential for real-time applications. * Specialized hardware will become more prevalent in AI. * Data cleaning is crucial to prevent the leakage of PII.