​(view in browser)​

Dear Reader,

Welcome to a belated Apr 8th issue of our newsletter! This weeks batch captures a wider shift from AI as spectacle to AI as engineering discipline: one piece makes quantization feel less like obscure systems trivia and more like the practical trick that lets serious models run smaller, cheaper, and closer to the edge; another pushes back hard on the fantasy that you can vibe your way to robust software without structure, testing, or taste; a third explores why language models can seem emotionally legible without actually “feeling” in any human sense; and taken together with the rise of personal agents and fresh warnings about crypto’s long-term exposure to quantum-era attacks, the message is clear that the next phase of machine learning will be shaped less by hype than by compression, interpretability, reliability, and security.

On the academic front, this weeks contributions suggest that progress in AI and science increasingly depend on seeing hidden structure clearly rather than treating outputs as self-explanatory: in code generation, that means making AI-assisted analysis more transparent, auditable, and easier to learn from, while even surprisingly simple self-distillation techniques can sharpen performance without exotic complexity; in language technology, it means recognizing that the cost of using large models is unevenly distributed across languages, with tokenization quietly shaping who pays more to participate; and beyond AI, the same lesson appears in studies of public health, brain science, and ecology, where policy timing can ripple into mortality, vaccination data can cut through speculation, psychedelic effects seem to reorganize large-scale brain circuits, and the resilience of plant-pollinator systems turns out to depend not just on who is connected to whom, but on when those connections happen.

Our current book recommendation is "Agentic Design Patterns: A Hands-On Guide to Building Intelligent Systems" by A. Gullí. You can find all the previous book reviews on our website. In this week's video, we have a tutorial on how to Deploy Agents with A2A on LangSmith Deployment.

Data shows that the best way for a newsletter to grow is by word of mouth, so if you think one of your friends or colleagues would enjoy this newsletter, go ahead and forward this email to them. This will help us spread the word!

Semper discentes,

The D4S Team

A. Gullí’s "Agentic Design Patterns: A Hands-On Guide to Building Intelligent Systems" feels like a timely guide for data scientists and machine learning engineers who are ready to move past the hype around AI agents and focus on how these systems are actually built. What makes the book stand out is its practical, pattern-based approach: instead of treating agents like magic, Gullí breaks them into reusable design ideas that help readers think more clearly about architecture, workflows, and implementation. That alone makes it more valuable than many AI books that are heavy on buzzwords and light on substance.

One of the book’s strongest qualities is its hands-on mindset. By working through recognizable frameworks and concrete design patterns, it gives technical readers a clearer path from experimentation to real system design. For ML engineers, that means a stronger grasp of modularity and maintainability; for data scientists, it offers a useful bridge between model knowledge and application building. The book is at its best when it helps readers see agentic systems not as mysterious novelties, but as engineering problems that can be approached systematically.

Its weaknesses are relatively minor but worth noting. Because it leans on current frameworks and tools, some parts may age quickly in such a fast-moving field, and readers looking for a deeper dive into evaluation, benchmarking, or production-scale operations may find it less comprehensive on those fronts. Still, Agentic Design Patterns sounds like the kind of book that can sharpen how technical practitioners think about intelligent systems—and for many readers, that will be reason enough to keep turning the pages.

1. ​Quantization from the ground up [ngrok.com]
2. ​Who Is Satoshi Nakamoto? My Quest to Unmask Bitcoin’s Creator [www.nytimes.com]
3. ​The Cult Of Vibe Coding Is Insane [bramcohen.com]
4. ​The Evolving Foundations of Math [quantamagazine.org]
5. ​Emotion Concepts and their Function in a Large Language Model [transformer-circuits.pub]
6. ​OpenClaw: The complete guide to building, training, and living with your personal AI agent [lennysnewsletter.com]
7. ​Safeguarding cryptocurrency by disclosing quantum vulnerabilities responsibly [research.google]

• ​Towards transparency and knowledge exchange in AI-assisted data analysis code generation (R. Haase)
• ​Premature termination of unemployment benefits increased COVID-19 transmission and deaths in the USA​ (S. Park, K. M. Lee, J. S. Earle)
• ​An international mega-analysis of psychedelic drug effects on brain circuit function​ (M. Girn, M. K. Doss, L. Roseman, K. H. Preller, F. Palhano-Fontes, L. Pasquini, F. S. Barrett, P. Mallaroni, N. L. Mason, C. Timmermann, D. E. McCulloch, P. M. Fisher, B. S. Winston, F. Moujaes, F. Muller, M. E. Liechti, F. X. Vollenweider, J. G. Ramaekers, K. Kuypers, D. B. Araujo, O. Sporns, J. Siegel, N. Dosenbach, D. J. Nutt, R. L. Carhart-Harris, E. A. Stamatakis, D. Bzdok)
• ​Association between COVID-19 vaccination and sudden death in apparently healthy younger individuals: A population-based case-control study​ (H. Abdel-Qadir, H. A. Bhatt, S. Swayze, M. Paterson, D. T. Ko, D. N. Juurlink, J. C. Kwong)
• ​Embarrassingly Simple Self-Distillation Improves Code Generation (R. Zhang, R. H. Bai, H. Zheng, N. Jaitly, R. Collobert, Y. Zhang)
• ​Do All Languages Cost the Same? Tokenization in the Era of Commercial Language Models (O. Ahia, S. Kumar, H. Gonen, J. Kasai, D. R. Mortensen, N. A. Smith, Y. Tsvetkov)
• ​Temporal Structure Mediates the Robustness and Collapse of Plant-Pollinator Networks (T. Clegg, T. Gross)

Deploy Agents with A2A on LangSmith Deployment

​

All the videos of the week are now available in our YouTube playlist.

​Unsubscribe​