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.
This conversation explores the shift in enterprise software as AI agents become primary users, moving towards 'headless' architectures where data and logic, rather than UIs, hold key value. It delves into the enduring stickiness of systems like SAP, driven by deep business process codification, and highlights that 'exception handling' is central to enterprise complexity. The discussion identifies new startup opportunities arising from AI, focusing on enhancing existing systems and bridging organizational functions rather than direct competition.
Benjamin Todd provides an updated career strategy for the AI era, explaining why 'follow your passion' is flawed and what truly builds fulfillment. He details the ABZ framework for planning under deep uncertainty and the 'moving bottleneck' concept to stay valuable as AI advances. The discussion highlights that a human-level digital worker quickly becomes superhuman and maps critical AI risks including power-seeking AI, extreme power concentration, and engineered pandemics, emphasizing that individual careers can be a powerful lever for good in these transformative times.
The Microsoft Research India Academic Research Summit 2026 session on "Inclusive AI" explored critical challenges in developing AI that serves diverse linguistic and cultural contexts. Speakers Niloy Ganguly, Danish Pruthi, Sunayana Sitaram, Anoop Kunchukuttan, and Ashutosh Modi addressed data gaps, model biases, and evaluation shortcomings, emphasizing the need for equitable and culturally relevant AI. Key themes included the use of synthetic data for low-resource languages, the impact of tokenization on model performance, geographical disparities in generative AI, and the application of AI for social good in legal and accessibility domains. The discussions underscored the importance of community involvement, open data, and designing AI for multilinguality from the outset, rather than as an afterthought.
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.
Jesse Zhang, co-founder and CEO of Decagon, discusses how their AI agents are revolutionizing customer service for large enterprises by replacing mundane human labor. He covers their go-to-market strategy, the importance of a hardworking in-office culture, his journey as a second-time founder, and the future of an agentic world where AIs interact on behalf of companies and consumers.
Learn how W&B Weave's online evaluations enable real-time monitoring of AI applications in production, allowing teams to track performance, catch failures, and iterate on quality over time using LLM-as-a-judge scores.
This summary covers a deep dive into the paper "A Survey of Context Engineering for Large Language Models". The discussion reframes the conversation from simple prompt engineering to a more systematic approach of building information environments for LLMs. It explores the foundational components of context engineering—generation, processing, and management—and their application in advanced systems like Retrieval-Augmented Generation (RAG), memory, tool use, and multi-agent systems.
Tudor Girba introduces "Moldable Development," a paradigm where developers create thousands of custom tools to analyze and interact with their specific software context. He demonstrates this live using Glamorous Toolkit, transforming an unstructured text file into a browsable system with custom views, all within minutes.
Experts from IBM X-Force discuss the alleged retirement of the Scattered Lapsus$ Hunters cybercrime gang, the ethics and implications of AI-powered ransomware, critical software supply chain vulnerabilities exposed by the recent npm hack, growing threats to Operational Technology (OT), and the emergence of AI-driven hiring fraud.
Edwin Chen, founder and CEO of Surge AI, shares the company's origin story, its rapid, bootstrapped growth, and its research-driven philosophy on data. He critiques traditional data labeling, explains why metrics like inter-annotator agreement fail for complex tasks, and offers a sharp analysis of benchmark hacking. Chen also details the future of data, from multimodal and agentic reasoning in rich RL environments to the need for hyper-specialized expertise for scientific discovery.
Get updates straight to your inbox. No spam, just useful content.