Things I read, watch, and listen to — links worth sharing.

• 
  understandingai.org

  Why Anthropic believes its latest model is too dangerous to release

  “The language models we have now are probably the most significant thing to happen in security since we got the Internet.”
  April 13, 2026
• https://erik-munkby.medium.com/your-ai-agent-doesnt-care-about-your-codebase-78e0338c40ef
  April 12, 2026
• christophermeiklejohn.com

  Software Engineering Is Becoming Civil Engineering
  April 12, 2026
• 
  Simon Willison’s Weblog

  Anthropic’s Project Glasswing—restricting Claude Mythos to security researchers—sounds necessary to me

  Anthropic didn’t release their latest model, Claude Mythos (system card PDF), today. They have instead made it available to a very restricted set of preview partners under their newly announced …
  April 12, 2026
• 
  Simon Willison’s Weblog

  Meta’s new model is Muse Spark, and meta.ai chat has some interesting tools

  Meta announced Muse Spark today, their first model release since Llama 4 almost exactly a year ago. It’s hosted, not open weights, and the API is currently “a private API …
  April 12, 2026
• 
  share.transistor.fm

  Engineering Enablement by DX | Measuring AI impact, assessing readiness, and new data trends

  In this episode of Engineering Enablement, Jesse Adametz joins Abi Noda, this time to host. Together, they explore how AI is showing up across the SDLC, not just in code generation, and how it is shifting bottlenecks across the development process. They unpack what “AI readiness” actually means...
  April 11, 2026
• 
  newsletter.pragmaticengineer.com

  DHH’s new way of writing code

  David Heinemeier Hansson shares why he shifted to an agent-first AI workflow, and what it means for how software is built and who builds it.
  April 10, 2026
• 
  lethain.com

  The agentic passive voice.

  At some point, you will have learned about the passive voice, where the actor in a sentence is unclear. For example, my software didn’t compile. That’s a good example of the passive voice. However, you might not know the full set of rules, because here are some sentences in the passive voice that you might not recognize: Claude made an error in my writeup. ChatGPT messed up the commitment. Gemini didn’t write tests. You might think those are active sentences, but those are in fact examples of the agentic passive voice. The rule here is: whenever the actor in a sentence is a model, then it’s a passive sentence. I’m sorry if your grammar instructor never taught you this rule, but this is just the way it works now.
  April 10, 2026
• 
  nytimes.com

  Hard Fork

  Each week, journalists Kevin Roose and Casey Newton explore and make sense of the rapidly changing world of tech.
  April 10, 2026
• 
  Westenberg.

  "Collaboration" is bullshit.

  This newsletter is free to read, and it’ll stay that way. But if you want more - extra posts each month, access to the community, and a direct line to ask me things - paid subscriptions are $2.50/month. A lot of people have told me it’s
  April 10, 2026
• 
  linkedin.com

  The Dark Knight Value: What Real Consulting Looks Like

  Why the best consultants aren't heroes. They're the ones willing to be the villain.
  April 10, 2026
• 
  newsletter.pragmaticengineer.com

  Scaling Uber with Thuan Pham (Uber’s first CTO)

  Thuan Pham (Uber's first CTO) on scaling Uber from constant outages to global infrastructure, the shift to microservices and platform teams, and how AI is reshaping engineering.
  April 9, 2026
• 
  GitHub

  GitHub - olgasafonova/tilbudstrolden-mcp: MCP server for Nordic grocery deal hunting via the Tjek API (Denmark, Norway, Sweden)

  MCP server for Nordic grocery deal hunting via the Tjek API (Denmark, Norway, Sweden) - olgasafonova/tilbudstrolden-mcp
  April 9, 2026
• 
  lennysnewsletter.com

  An AI state of the union: We’ve passed the inflection point, dark factories are coming, and automation timelines | Simon Willison

  Listen now | Simon Willison on why November 2025 changed software engineering forever, the lethal trifecta, his top agentic engineering patterns, and much more
  April 9, 2026
• 
  shows.acast.com

  46. AI, agile och vattenfall m. Jakob Wolman | Datastudion

  Listen to 46. AI, agile och vattenfall m. Jakob Wolman from Datastudion. Jakob Wolman – mjukvaruutvecklare, agile coach och engineering manager med rötter i Malmö och Köpenhamn – gästar Datastudion för ett samtal som startade i ett LinkedIn-inlägg av Benny.Frågan: när kostnaden för att skriva kod går mot noll, vad händer med det agila arbetssättet? Plötsligt verkar det logiskt att lägga all energi på kravspecen igen – och låta AI sköta implementationen. Men är det verkligen vattenfall, eller är det något nytt?Vi landar i feedback loops, flaskhalsar som flyttar sig ett steg, Lean-tankar om att återinvestera frigjord tid – och varför marshmallow-utmaningen avslöjar mer om hur vi egentligen tänker än vad något agilt ramverk gör.ReferenserJakob Wolman – mjukvaruutvecklare, agile coach, engineering managerØredev – teknikkonferens i MalmöRUP – Rational Unified Process, strukturerad mjukvarumetodikVattenfall/Waterfall – sekventiell projektmetodikAgile – iterativ mjukvaruutvecklingLean – minimera slöseri, maximera flödeTheory of Constraints – identifiera och hantera flaskhalsarKanban – visuell metod för arbetsflödeBeyond Budgeting – rörelse mot mer adaptiv budgeteringSusanne Kaiser – föreläsare och författare om flow och mjukvaruorganisationThe Marshmallow Challenge – övning i iterativt tänkande och teamdynamikNobody ever gets credit for fixing problems that never happen – Repenning & Sterman, artikel om kvalitet och återinvesteringSpec-driven development – kravspecifikation som driver implementationClaude Code – Anthropics CLI-verktyg för agentisk kodningMCP – Model Context Protocol – standard för att koppla AI-agenter till externa verktygKordsnack – podd om IT, nämns i samband med COBOL-avsnittCOBOL – programspråk vanligt inom mainframe, särskilt bankerMainframe – storskaliga datorsystem, fortfarande i bruk inom finansBill Gates TED Talk – om noll vs nästan noll (referens till skillnad på noll och nästan noll)Nick Borgen – World Wide Web – låt nämnd som historisk parallell till AI-hypen
  April 9, 2026
• 10 Claude Code Principles

  10 Claude Code Principles

  Most AI coding advice is wrong. Not “slightly suboptimal” wrong — measurably, scientifically, provably wrong.
  April 7, 2026
• 
  tidyfirst.substack.com

  Starving Genies

  What 3X: Explore/Expand/Extract says about throttling
  April 7, 2026
• 
  elenaverna.com

  Confessions of a Millennial in Tech

  A lot of mixed feelings and unanswered questions.
  April 7, 2026
• 
  DR LYD

  Prompt | Sociale mediers tobacco-moment og Pokémon no-go | Lyt som podcast | DR LYD

  Lyt som podcast her på DR LYD. Det viser sig, at der muligvis lå en hemmelig plan bag Pokémon Go: at bruge spillernes data til at bygge et ekstremt præcist kort over den virkelige verden - en slags digital tvilling, som for eksempel leveringsrobotter kan navigere efter. Der er altid en lidt skummel bagtanke, når det kommer til tech. Her var den bare pakket ind i nuttede Pokémoner. Nogle kalder det sociale mediers tobacco moment. Meta og YouTube er blevet dømt i en retssag for uagtsomt platformdesign - altså for at have skabt vanedannende produkter, som ifølge sagen var med til at føre til angst og depression. Hvorfor den dom potentielt kan få stor betydning, taler vi med psykolog og sagkyndigt vidne i retssagen, Lotte Rubæk, om. Vi omtaler i programmet selvmord. Hvis du er i krise eller har selvmordstanker, er det vigtigt, du taler med nogen. Kontakt din egen læge, lægevagten eller Livslinien, hvis du har brug for hjælp. Ring 1-1-2, hvis situationen er livstruende. Værter: Marie Høst, Tech-journalist og moderator; Henrik Moltke, Tech-journalist. Gæst: Lotte Rubæk psykolog med speciale i selvskade, og vidne i retssagen. I redaktionen: Buster Hoff.
  April 4, 2026
• 
  Audible.co.uk

  The Undoing Project

  Check out this great listen on Audible.com. Penguin presents the unabridged, downloadable, audiobook edition of The Undoing Project, by Michael Lewis, read by Dennis Boutsikaris.Daniel Kahneman and Amos Tversky met in war-torn 1960s Israel. Both were gifted young psychology professors: Kahneman a ...
  April 2, 2026
• 
  DN.se

  Kristofer Ahlström: Åtta skäl att förakta AI

  Det kallas billig smörja, dåliga produkter och anses utmana själva uppfattningen om mänsklighet. Varför väcker AI-skapat innehåll sådan vrede?
  March 29, 2026
• 
  freakonomics.com

  Freakonomics radio
  March 28, 2026
• 
  nytimes.com

  Opinion | How Fast Will A.I. Agents Rip Through the Economy? - The New York Times
  March 28, 2026
• 
  shows.acast.com

  45. Sömnbrist, oro och AI-assisterad kodning | Datastudion

  Listen to 45. Sömnbrist, oro och AI-assisterad kodning from Datastudion. Hugo är tillbaka från föräldraledigheten – och har spenderat en del av den med att doomscrolla på LinkedIn och oroa sig för att bli obsolet. Har han det? Benny satte sig ner en dag och vibe-kodade en Slackbot på 27 minuter. Det tog sedan tolv timmar till att den kändes okej. Han är fortfarande stolt.I det här avsnittet pratar vi om hur mjukvaruutveckling faktiskt ser ut nu – och om det verkligen är ett skifte eller om vi bara har fått ett bättre tangentbord. Vi går igenom CLAUDE.md, varför du bör byta fönster oftare än du tror, och varför det fortfarande inte är för sent att hoppa på tåget.Spoiler: det byter lok hela tiden.ReferenserVibe coding – kodning med AI-assistans där man beskriver vad man vill ha i fritextLLM – Large Language Model, stora språkmodeller som ChatGPT och ClaudeClaude Code – Anthropics CLI-verktyg för AI-assisterad utvecklingCLAUDE.md – instruktionsfil som styr hur Claude Code beter sig i ett projektVisual Studio Code – populär kodredigerareChatGPT – OpenAIs AI-assistentFlask – lättviktigt Python-webbramverkFastAPI – modernt Python-ramverk för API-utvecklingSlack API – Slacks API för att bygga bottar och integrationerRAG – Retrieval-Augmented Generation, teknik för att ge AI tillgång till extern informationSpec-driven development – utvecklingsmetodik där man specificerar beteende innan man kodarMVP – Minimum Viable Product, minsta möjliga fungerande produktSLP – Speledarens person, NPC i rollspelssammanhang
  March 28, 2026
• 
  Honeycomb

  Production Is Where the Rigor Goes

  In early February, Martin Fowler and the good folks at Thoughtworks sponsored a small, invite-only unconference in Deer Valley, Utah — birthplace of the Agile Manifesto — to talk about how software engineering is changing in the AI-native era.
  March 27, 2026
• 
  Thoughtworks

  The Future of Software Development Retreat

  Deer Valley, Utah, 2026 — an invite-only retreat exploring where software development is headed.
  March 27, 2026
• 
  Susanne Kaiser

  Building Foundations for Continuous (AI-accelerated) Change

  How organizations can build the structural and cultural foundations needed to sustain continuous change in an AI-accelerated world.
  March 27, 2026
• 
  The New York Times

  Hard Fork: AI Washing, Tokenmaxxing

  The Hard Fork podcast explores the latest in tech — from companies overstating their AI capabilities to the new strategies around crypto tokens.
  March 27, 2026
• 
  Lenny's Newsletter

  The art of influence: The single most important skill that AI can't replace

  A PM's guide to influencing executives: align with their incentives, present three options, and follow the breadcrumbs. With Jessica Fain (Webflow, ex-Slack).
  March 27, 2026
• 
  newsletter.getdx.com

  AI productivity gains are 10%, not 10x

  Preliminary data from our longitudinal AI impact study
  March 12, 2026
• 
  Medium

  Welcome to Gas Town

  Happy New Year, and Welcome to Gas Town!
  March 11, 2026
• 
  Medium

  Welcome to the Wasteland: A Thousand Gas Towns

  Howdy partner, and welcome to the Wasteland!
  March 11, 2026
• 
  Coinerella

  "Made in EU" - it was harder than I thought.

  I tried building my startup entirely on European infrastructure. Here's the stack I landed on, what was harder than expected, and what you still can't avoid.
  March 9, 2026
• Mitchell Hashimoto

  My AI Adoption Journey
  February 24, 2026
• 
  Management Excellence by Art Petty

  Five big mistakes to avoid when changing careers

  An important part of any career shift process is minimizing risk. Here are five big mistakes to avoid if you are thinking about a change.
  February 24, 2026
• 
  YouTube

  FULL DISCUSSION: Yuval Noah Harari Warns AI Will Take Over Language, Law, and Power at WEF | AI1G

  Renowned historian Yuval Noah Harari delivers a chilling warning at World Economic Forum 2026, arguing that AI is no longer a tool but an agent that can thin...
  February 19, 2026
• 
  agglomerations.substack.com

  AI and the Economics of the Human Touch

  A reason for optimism
  February 18, 2026
• 
  linkedin.com

  Ready Player None

  Last year I spent five posts exploring what happens when AI breaks the consulting model. The pyramids crumble.
  February 17, 2026
• 
  open.substack.com

  The 70/30 Flip: What AI Is Actually Changing About Work

  Why your resume doesn't capture what makes you valuable anymore
  February 17, 2026
• 
  omny.fm

  Starting 2026 - Unsupervised Learning

  A welcome back and early entry into 2026.  Sponsored by: Knocknoc!
  February 4, 2026
• 
  lethain.com

  "Good engineering management" is a fad

  As I get older, I increasingly think about whether I’m spending my time the right way to advance my career and my life. This is also a question that your company asks about you every performance cycle: is this engineering manager spending their time effectively to advance the company or their organization? Confusingly, in my experience, answering these nominally similar questions has surprisingly little in common. This piece spends some time exploring both questions in the particularly odd moment we live in today, where managers are being told they’ve spent the last decade doing the wrong things, and need to engage with a new model of engineering management in order to be valued by the latest iteration of the industry.
  December 5, 2025
• 
  Bloomberry

  I analyzed 180M jobs to see what jobs AI is actually replacing today

  Last Updated: March 14, 2026What impact is AI having on the job market? Everyone has an opinion, but…
  November 25, 2025
• 
  LinkedIn

  #thanks | Evyatar Bluzer | 11 comments

  I was tired of consulting with multiple LLMs. So I vibe-coded this side project - a tool to automatically compare answers from multiple AI models for questions I really care about (think: health, finance, kids' education). Started it as something just for myself, then I wrapped it as a web app so others could try it too. Under the hood, it uses some principles from research on improving AI reliability (details in first comment👇). Would love for you all to check it out and let me know what you think, or how you’d use it! https://trustedchat.chat #thanks Ilan Voronel for the support and expert advise! | 11 comments on LinkedIn
  November 24, 2025
• Amazon Web Services, Inc.

  Summary of the Amazon DynamoDB Service Disruption in the Northern Virginia (US-EAST-1) Region
  November 11, 2025
• 
  LinkedIn

  Novo Nordisk har lige slået stillingen som Chief Artificial Intelligence Officer (CAIO) op, og jobopslaget er vanvittigt interessant, hvis man ser på det med AI adoptions- og transformationsbriller.… | Dan Rose Johansen | 61 comments

  Novo Nordisk har lige slået stillingen som Chief Artificial Intelligence Officer (CAIO) op, og jobopslaget er vanvittigt interessant, hvis man ser på det med AI adoptions- og transformationsbriller. Opslaget giver indsigt i, hvilke ansvar og kvalifikationer de vægter, og det giver et indblik i enterprise virksomheders tilgang til AI. Der er meget positivt, men også et par centrale valg, der er værd at diskutere. 1) En bred og strategisk rolle Novo Nordisk har tydeligvis set, at det her er en BRED og transformerende rolle. Der er et kæmpe fokus på ledelse, organisatorisk transformation og tværgående relationer. Det er så fedt at se! Alt for mange virksomheder isolerer deres AI-ledere i en snæver, teknisk rolle, der i praksis gør alle indsatser taktiske i stedet for strategiske. Her sigter Novo Nordisk højere. 2) Ejerskabet Opslaget signalerer et centralt AI-ejerskab. CAIO'en skal "eje" og "drive" udrulningen og transformationen. Det er interessant, da best practices peger på, at ejerskab skal decentraliseres og ligge ude i forretningsfunktionerne. CAIO-funktionen skal agere support og enabler. Gør man det for centralt, er der en kæmpe risiko for, at initiativerne sander til og aldrig bliver forankret. 3) De tekniske kvalifikationer På trods af det brede fokus på transformation og ledelse, leder de stadig efter en profil med en tung, teknisk uddannelse. For mig er det et lille levn fra "AI er et sæt tekniske værktøjer"-tiden. Er den bedste til at drive kompleks forandringsledelse og organisationsudvikling nødvendigvis den samme person som den med den dybeste tekniske baggrund? Det er jeg ikke sikker på. 4) Placeringen Rollen er placeret i Enterprise IT. Det er helt klassisk, men ikke helt godt. Reglen er simpel: Hvis AI ikke diskuteres dér, hvor strategien sættes (i direktionen), bliver AI aldrig ægte strategisk. Det risikerer at fastholde AI som en IT-disciplin frem for en kerneforretnings-disciplin. Jeg håber Novo Nordisk finder den rigtige, for rollen er virkeligt afgørende for den næste tid. | 61 comments on LinkedIn
  November 11, 2025
• 
  newsletter.pragmaticengineer.com

  State of the software engineering jobs market, 2025: what hiring managers see

  Observations by 30+ hiring managers and tech recruiters about what’s happening: a flood of inbound applications means more selective hiring, there’s increased demand for product engineers, and more
  October 8, 2025
• 
  newsletter.pragmaticengineer.com

  How tech companies measure the impact of AI on software development

  How do GitHub, Google, Dropbox, Monzo, Atlassian, and 13 other companies know how well AI tools work for devs? A deepdive sharing exclusive details, with CTO Laura Tacho
  September 23, 2025
• 
  anthropic.com

  A postmortem of three recent issues

  This is a technical report on three bugs that intermittently degraded responses from Claude. Below we explain what happened, why it took time to fix, and what we're changing.
  September 22, 2025
• 
  Amazon Science

  Measuring the effectiveness of software development tools and practices

  New cost-to-serve-software metric that accounts for the full software development lifecycle helps determine which software development innovations provide quantifiable value.
  September 17, 2025
• 
  Learning Loop

  Hedging against AI

  Why simple card decks keep your brain in the game
  September 8, 2025
• 
  cutlefish.substack.com

  TBM 377: Time Allocation ≠ Capacity Allocation

  Spoiler: the concept of capacity allocation is fraught with misunderstanding, and companies waste a lot of money (and do a lot of damage) by mistaking time allocation for capacity allocation.
  September 8, 2025
• 
  jamesjboyer.substack.com

  On Being Liked or "Mistaking Harmony for Health"

  The difference between good vibes and good leadership.
  June 26, 2025
• 
  charity.wtf

  In Praise of “Normal” Engineers

  This article was originally commissioned by Luca Rossi (paywalled) for refactoring.fm, on February 11th, 2025. Luca edited a version of it that emphasized the importance of building “10x engi…
  June 26, 2025
• 
  newsletter.pragmaticengineer.com

  Learnings from two years of using AI tools for software engineering

  How to think about today’s AI tools, approaches that work well, and concerns about using them for development. Guest post by Birgitta Böckeler, Distinguished Engineer at Thoughtworks
  June 26, 2025