This paper challenges the prevailing "bigger is better" narrative in AI, arguing that Small Language Models (SLMs) are not just sufficient but often superior for agentic AI tasks due to their efficiency, speed, and specialization. The discussion explores the paper's core arguments, counterarguments, and the practical implications of adopting a hybrid LLM-SLM approach.
Nikolaos Vasiloglou, VP of Research ML at RelationalAI, shares his extensive analysis of the 2023 NeurIPS conference, distilling over 200 hours of content. Key themes include the dominance and evolution of agentic AI, the state of open-source vs. frontier LLMs, the first signs of deep learning models outperforming XGBoost on tabular data, and the critical rise of verification systems. He also explores the future of AI with data attribution for monetization and the concept of composable, LEGO-like language models.
Ben Kus, CTO of Box, outlines the technical evolution of their AI platform, detailing the transition from a promising but fragile LLM-based metadata extraction system to a robust, scalable agentic architecture. He explains why this shift was necessary to handle enterprise-level complexity and the key lessons learned.
An exploration of designing AI-native user experiences by treating AI systems as "coworkers." This summary covers a framework for AI interaction (invisible, ambient, inline, conversational), a design philosophy based on "feeling the material" and "courting emergence," and novel UX patterns for collaborative AI tools.
Andrew Ng discusses the rise of agentic AI, moving beyond scale as the sole driver of progress. He explores how AI-assisted coding is creating a new startup paradigm, shifting the bottleneck from engineering to product management and favoring technical founders. Ng argues for smaller, highly-skilled teams and predicts AI will profoundly empower individuals across all job functions.
Explore the concept of "self-driving storage," where AI and AIOps autonomously manage data infrastructure. Learn how mobile storage partitions, predictive analytics, and agentic AI are used to automate capacity management, workload placement, and on-demand performance optimization without human intervention.