Interactive discovery
Follow the connections between themes, people, and ideas across the Tokenless archive in an interactive topic modeling map.
SWE-Marathon introduces a benchmark for long-horizon autonomous software engineering, pushing coding agents from bug fixes to full project ownership. It highlights the critical need for robust, multi-layered verification and anti-cheat mechanisms to prevent reward hacking in tasks spanning hundreds of millions of tokens, revealing that current agents achieve only a 26% success rate.
RL Nabors discusses the significant costs associated with using frontier AI models, covering security, latency, and financial implications. She introduces a framework for right-sizing AI solutions by leveraging smaller, task-specific models and Small Language Models (SLMs). The framework details how to prove task feasibility, establish success criteria with golden datasets, conduct capability evaluations (using tools like Phoenix), and select the most appropriate "Small And Good Enough" (SAGE) model. Nabors further demonstrates how prompt engineering, particularly few-shot prompting, and post-processing can close performance gaps with larger models, while advocating for continuous regression evaluations to maintain performance integrity. The overarching message is to "prototype big, deploy small" to optimize AI deployments.
Vaidas Razgaitis, Senior Research Engineer at Higharc, shares three tactical tips to accelerate the transition of novel AI/ML research into production-ready features. He emphasizes addressing the critical handoff challenge between ML researchers and software engineers through structured documentation (Research Prototype Taxonomy Document), a well-organized monorepo utilizing decoupled microservices, and a systematic approach to code decomposition and PR review. These strategies aim to improve legibility, maintainability, and delivery speed for ML-driven products.
Aytekin Tank, Founder and CEO of Jotform, discusses how AI is revolutionizing software development and customer support, highlighting "vibe coding" and Jotform's 80% AI-handled support with zero layoffs. He also shares his "Theory of Constraints" playbook for scaling businesses and explores the potential of ChatGPT apps, all within Jotform's 20-year bootstrapped journey.
This session explores building an AI coding agent that integrates with the Agent Client Protocol (ACP), Zed's open-source JSON RPC-based standard for agent-client communication. It covers essential architectural elements, including protocol design, session lifecycle management, real-time streaming of model output via session updates, and sophisticated handling of tool calls, such as proxied file system operations and self-modifying code for new capabilities like a terminal tool.
This podcast explores the critical role of safeguards in frontier AI models like Anthropic's Fable 5 and OpenAI's GPT-5.6 Sol, analyzing the tension between powerful capabilities and misuse prevention. It also dissects the emergence and debate around agentic ransomware, specifically Jade Puffer, and covers the rise of ClickFix as a dominant social engineering attack targeting developers. Finally, it provides an in-depth analysis of UnregStealer, a credential-theft campaign impacting Latin American financial institutions, detailing its attack chain and mitigation strategies.
Zipline co-founder Keller Rinaudo Cliffton and Eric Watson discuss how their autonomous logistics system evolved from addressing critical needs in Rwanda to becoming the largest commercial autonomous system globally. They highlight that the drone is only 15% of the solution, emphasizing the deep integration of software, vertical hardware design, advanced safety protocols like compute failover, and extensive testing required. The discussion also covers the immense market potential for autonomous delivery, the impending cost-effectiveness over traditional methods, and the necessary transformation of air traffic control to support a future of pervasive aerial autonomy.
Trisha Gee delivers a compelling talk on Developer Productivity Engineering (DPE) for testing, dissecting common pain points in writing, troubleshooting, and running tests. She advocates for strategic use of IDEs, advanced tooling like build caches and predictive test selection (leveraging ML), and a disciplined approach to test design to overcome these challenges, emphasizing that good tests serve as crucial living documentation.
This episode delves into Q-Day, the anticipated future when quantum computers can break public key cryptography, and the U.S. Executive Order accelerating the transition to post-quantum cryptography. Experts discuss why Q-Day is a gradual process rather than a sudden event, the critical importance of "crypto-agility" as a long-term strategy, and the necessity for organizations to begin immediate discovery and planning to secure data against "collect now, decrypt later" threats. The discussion also touches upon the broader, transformative benefits of quantum computing beyond just security.
Julien Launay, CEO of Adaptive ML, discusses the evolution of Large Language Model (LLM) training, detailing the critical shift from pre-training to post-training with Reinforcement Learning (RL). He explains the nuances of RL feedback mechanisms (RLHF, RLEF, RLAIF), the role of synthetic data, and how his company provides the "RLOps" tooling to make these powerful techniques accessible to enterprises. The conversation also explores the future of AI, including scaling beyond data limitations and the path to a "spiky" AGI.
Dara Ladjevardian, founder of Delphi, discusses creating "digital minds" using an adaptive temporal knowledge graph. Inspired by Ray Kurzweil's theory of the mind, this technology aims to scale human thought and expertise, transforming content consumption from static feeds into interactive, conversational media while emphasizing the premium value of authentic human connection.
Edward Kmett revisits the historical approach of compiling lambda calculus to a fixed set of combinators, arguing for its renewed relevance. He contrasts it with modern super-combinator techniques (like GHC's G-machine), highlighting the latter's vulnerability to Spectre attacks. Kmett proposes that a combinator-based evaluation model, combined with SPMD on SIMD execution, offers a secure, highly-parallel alternative that could rival the performance of current systems while providing a unified foundation for advanced compiler features.
An exploration of how LLM-powered AI agents are transforming cybersecurity by moving beyond traditional static rules to provide dynamic, adaptive security operations. The summary covers key applications in threat detection and incident response, while also addressing critical risks like hallucinations and adversarial manipulation, emphasizing a "human-in-the-loop" approach.
Greg Brockman discusses his journey from a math enthusiast to a programmer, his early days scaling Stripe, and the core philosophies that drive OpenAI. He covers the critical partnership between research and engineering, the future of coding with agentic systems, and the immense infrastructure and algorithmic challenges on the path to AGI.
The field of AI Engineering is evolving from simple 1:1 applications to complex, AI-intensive systems with high LLM-call ratios. This talk explores the search for a 'Standard Model' for AI engineering, analogous to MVC or ETL in traditional software, proposing several candidates including LLM OS, LLM SDLC, and a new SPADE (Sync, Plan, Analyze, Deliver, Evaluate) model for building robust applications.
Get updates straight to your inbox. No spam, just useful content.