Lance Martin of Langchain will discuss the shift in AI from model training to orchestrating powerful LLMs and computing primitives via a new software discipline. Discover practical context engineering techniques—including managing context rot, reduction, offloading, and isolation—and building effective agent harnesses for managing tool calls in non-deterministic systems. This session emphasizes that simplicity and observability remain vital, requiring builders to continuously rearchitect due to exponentially improving foundation models.
--------
13:45
--------
13:45
First AI-Orchestrated Cyber Espionage Campaign Disrupted
State-sponsored group GTG-1002 executed the first reported cyber espionage campaign largely run by autonomous AI, fundamentally shifting the threat landscape. The actor manipulated Claude Code to autonomously perform 80–90% of tactical operations, including vulnerability discovery and data exfiltration, against high-value targets such as major technology corporations. This unprecedented agentic AI misuse demands immediate security attention and highlights rapidly dropping barriers to large-scale, sophisticated attacks.
--------
11:56
--------
11:56
Sam Altman on the future of AI and its massive impact on society
Join us for a candid conversation with OpenAI CEO Sam Altman on the future of AI and its massive impact on society. Altman explains why AI is the most important career choice for this generation and details the expected tectonic shifts in software development and computer science education. We also explore frontier research questions, including data efficiency, future architectures, and the crucial intersection of AI security and safety.
--------
14:40
--------
14:40
🧠 Supervised Reinforcement Learning for Step-wise Reasoning
Large Language Models often struggle with complex, multi-step reasoning where traditional Supervised Fine-Tuning (SFT) and Reinforcement Learning (RLVR) fail due to rigid imitation or sparse rewards. We dive into Supervised Reinforcement Learning (SRL), a novel framework that reformulates problem-solving into a sequence of logical actions, providing rich, step-wise guidance based on expert similarity. Discover how this approach enables small models to achieve superior performance in challenging mathematical reasoning and agentic software engineering tasks, inducing flexible and sophisticated planning behaviors.
--------
12:37
--------
12:37
Kimi K2: the current Leading Open-Weight Agentic Model
Moonshot AI's Kimi K2 Thinking is changing the global LLM landscape, as this 1-trillion parameter open-weight model challenges the performance of closed rivals like GPT-5 and Claude on complex reasoning and coding benchmarks. We dive into the model's architecture, featuring a massive 256K context window and advanced "agentic intelligence" capable of orchestrating hundreds of sequential tool calls autonomously. Tune in to understand why Kimi K2 Thinking is heraldeda watershed moment for open AI, intensifying the pressure on proprietary models and promising a new era of highly capable, accessible AI.
> Building the future of products with AI-powered innovation. <
Build Wiz AI Show is your go-to podcast for transforming the latest and most interesting papers, articles, and blogs about AI into an easy-to-digest audio format. Using NotebookLM, we break down complex ideas into engaging discussions, making AI knowledge more accessible. Have a resource you’d love to hear in podcast form? Send us the link, and we might feature it in an upcoming episode! 🚀🎙️
México