An expert overview of post-training techniques for language models, covering the entire workflow from data generation and curation to advanced algorithms like Supervised Fine-Tuning (SFT), Direct Preference Optimization (DPO), and Reinforcement Learning (RL), along with practical advice on evaluation and iteration.
This talk introduces Anaximander, a system designed to bridge the gap between traditional, GUI-driven Geographic Information System (GIS) workflows and modern, code-heavy machine learning practices. Anaximander integrates geospatial foundation models directly into QGIS, allowing experts to interactively orchestrate, run, and evaluate models for tasks like semantic segmentation and object detection on satellite imagery.
At Applied Compute, efficient Reinforcement Learning is critical for delivering business value. This talk explores the transition from inefficient synchronous RL to a high-throughput asynchronous 'Pipeline RL' system. The core challenge is managing 'staleness'—a side effect of in-flight weight updates that can destabilize training. The speakers detail their first-principles systems model, based on the Roofline model, used to simulate and find the optimal allocation of GPU resources between sampling and training, balancing throughput with algorithmic stability and achieving significant speedups.
a16z co-founders Marc Andreessen and Ben Horowitz discuss the shift to a decentralized media ecosystem, their investment thesis on supply-driven markets, and the transformative impact of AI. They detail the a16z model of leveraging reputation as a core asset to turn inventors into CEOs and explain why AI represents a fundamental reinvention of computing that will unlock unprecedented growth.
Ben Luria, CEO of Hirundo, discusses the critical need for machine unlearning, framing it as a form of "AI neuro-surgery" for enterprise AI. He explains how this technique directly modifies model weights to remove unwanted data and behaviors, addressing core risks that superficial solutions like guardrails cannot solve.
State Space Models (SSMs) are emerging as a powerful and efficient alternative to Transformers for handling sequential data. Aaron Baughman explains the core concepts of SSMs, their mathematical foundations, and how architectures like S4 and Mamba address the memory and scalability challenges inherent in Transformers, leading to a new generation of faster, more intelligent hybrid AI models.
Nikesh Arora, CEO of Palo Alto Networks, shares his unconventional journey and leadership philosophy. He provides a masterclass in building a multi-platform company through strategic M&A, explains why founders should sometimes ignore customers, and reveals how to lead with conviction while managing imposter syndrome.
Stewart Butterfield, co-founder of Slack and Flickr, shares the product frameworks and leadership principles that guided his success. He delves into concepts like "utility curves" for feature investment, the "owner's delusion" in product design, and why focusing on "comprehension" is often more important than reducing friction. He also introduces powerful mental models for organizational effectiveness, such as combating "hyper-realistic work-like activities" and applying Parkinson's Law to team growth.
Intuit CEO Sasan Goodarzi discusses the operational playbook for reinventing a 40-year-old company, from its slow transition to SaaS to its early adoption of AI. He shares insights on winning the SMB market by treating small businesses like consumers, building effective channel partnerships, and developing a platform strategy. Goodarzi also details his leadership philosophy, emphasizing that grit and curiosity are more critical than raw talent.
Michael Truell, co-founder of Cursor, shares the journey of building an AI-first code editor that grew to $100M ARR in a year. He discusses the crucial pivot from failed startup ideas, the conviction required to compete with GitHub Copilot, and the lessons learned in product development, model strategy, and the future of programming in a world where software development will increasingly flow through models.
The video explains the distinct but complementary roles of AI governance and security in mitigating AI risks. It contrasts their focuses, from self-inflicted policy violations (governance) to intentional external attacks (security), and proposes a layered framework combining both for comprehensive protection.
OpenAI's Build Hour provides a deep dive into GPT-5, showcasing its advanced coding and agentic capabilities. The session covers the new Responses API, critical for leveraging the model's reasoning, along with new parameters for steerability and practical prompting techniques for building complex, reliable applications.
Built-in tools like web search, file search, and code interpreter allow developers to extend model capabilities out-of-the-box. This summary covers the concepts, compares them to function calling, and details a demo of building a data exploration dashboard using multiple tools in concert.
A deep dive into building sophisticated voice agents using OpenAI's Realtime API and Agents SDK. The session covers architectural patterns like chained vs. end-to-end models, the use of multi-agent systems with handoffs for specialized tasks, and best practices for production including debugging with traces, implementing guardrails, and creating robust evaluations.
A deep dive into Reinforcement Fine-Tuning (RFT), covering how to set up tasks, design effective graders, and run efficient training loops to improve model reasoning, based on a live demonstration from OpenAI's Build Hours.
Get updates straight to your inbox. No spam, just useful content.