Sign in

kimrhh

@kimrhh.bsky.social
206 followers 1K following 22 posts

He/Him. Sailing 🤩 DevOps with passion and opinions. All things agile. proud father of two crazy teen girls. Love the people part of SW dev.

PostsRepliesMedia
Reposted by kimrhh
Dylan Beattie @dylanbeatt.ie · 25/09/2026
My workstation is a Ryzen 9050 with 128Gb of RAM, an RTX 5070 and three NVMe drives, and it is running Adobe Media Encoder like a very tired tortoise trying to suck treacle through a paper straw. Could we stop dicking about with AI models and get back to making computers faster and betterer please?
3171
Reposted by kimrhh
Chris Oldwood @chrisoldwood.bsky.social · 24/09/2026
The practice of sticking all C# extension methods in a top-level folder called Extensions needs to get in the bin 😔. Grouping code by the implementation mechanism is just weird. We don’t put all abstract classes in a top-level folder called Abstracts, or have a Statics folder for static classes. 🤷‍♂️
741
Reposted by kimrhh
Furkan Onder @furkanonder.bsky.social · 26/09/2026
Asahi Linux running on an M3 MacBook. No VMs, no hacks... just `GNU/Linux` on Apple Silicon. #AsahiLinux #FreeSoftware
152
Reposted by kimrhh
Barbara Forbes @ba.4bes.nl · 24/09/2026
Git Tip #27: You made two unrelated changes in the same file. How do you commit just one of them? Find my new short on Youtube: www.youtube.com/@Ba4bes/shorts #git #softwaredevelopment #devtips
111
Reposted by kimrhh
Gergely Orosz @gergely.pragmaticengineer.com · 22/09/2026
A look into how the Windows team is integrating AI into their OS: eg adding agent identities, akin to that of a user. It's the first opinionated approach I see for AI agents at the OS level - and surely other OSes will do sg similar. Deepdive: newsletter.pragmaticengineer.com/p/windows-an...
1253
Reposted by kimrhh
Frank Wiles @fwiles.bsky.social · 22/09/2026
A couple of weeks ago I gave a short talk on Open Source Maintainership in an LLM world, I turned it into a blog post because I wanted to be able to share the advice to a wider audience. frankwiles.com/posts/open-...
frankwiles.com
Open Source Maintainership in an LLM world
Suggestions and techniques you can use to keep your sanity in a sea of issues and PRs
143
Reposted by kimrhh
Aaron Powell (@slace) @aaron-powell.com · 23/09/2026
I wrote a blog for @visualstudio.com on just how easily you can use the Parallel Stacks view to start debugging memory dumps from production systems - devblogs.microsoft.com/visualstudio... There's a companion blog on the .NET blog about how to write a dump collector.
devblogs.microsoft.com
Today I will... debug a production crash - Visual Studio Blog
Learn effective debugging techniques for C# applications and how to tackle performance issues with memory dumps.
022
Reposted by kimrhh
Laurent Kempé 🚀 @laurentkempe.com · 19/09/2026
What if AI returned decisions your C# code could use directly, without parsing JSON? 🤖 Explore TypeSafe AI Jev model and an unofficial .NET SDK for structured AI output. 📖 Blog post: laurentkempe.com/2026/09/19/t... #dotnet #csharp #AI #MachineLearning
laurentkempe.com
Laurent Kempé - TypeSafe's Jev AI Model in .NET: A Community SDK for Structured AI Output in C#
Most AI tooling I've used in the last couple of years follows the same pattern. You send a prompt, get back text, and then write brittle code to turn that te...
011
Reposted by kimrhh
David Whitney @davidwhitney.co.uk · 18/09/2026
Small hobby project off the side of a hobby project. I'm doing some game preservation/restoration on GTA1. 32/64 bit support, Mac/Linux port, modernized controls and camera. Here's a native Mac port...
1101
Reposted by kimrhh
Adron @adron.bsky.social · 18/09/2026
Quick proposal: it's time to give up pull requests. PRs were built for a world where a human wrote code slowly, thought hard about every line, and needed a review gate before merge. That world's gone. 🧵 1/7
101
Reposted by kimrhh
Damian Edwards @damianedwards.com · 13/09/2026
Built a new thing to make installing & updating .NET on Windows easier after realizing I had *four* SDK installs to update on patch Tuesday: dotnet-steward.damianedwards.dev Install-PSResource DotNetSteward -Scope CurrentUser -TrustRepository non-Windows just use aka.ms/dotnetup
Screenshot of PowerShell terminal using the DotNet-Steward PowerShell module to manage .NET installations on Windows
1143
Reposted by kimrhh
T'otherAlistair @totheralistair.bsky.social · 29/08/2026
i feel that “psychological safety" is getting too much press these days, so I added this new assertion (that was missing, actually) While important, it's only one of a series of factors in "speaking up” & “getting heard” COmments, criticism, suggestions welcome
023
Reposted by kimrhh
Damien Bowden @damienbod.com · 17/08/2026
Blogged: Use Aspire to implement and deploy the BFF security architecture damienbod.com/2026/08/17/u... #aspnetcore #dotnet #aspire #oauth #oidc #auth0 #bff #iam #angular #owasp #sonar
damienbod.com
Use Aspire to implement and deploy the BFF security architecture
This blog demonstrates how to use Aspire to set up a solution for developing and deploying an ASP.NET Core web application with Auth0 as the identity provider and a downstream API. The application …
021
Reposted by kimrhh
Michael Kohl @citizen428.net · 17/08/2026
The weirdest thing to me about #vibecoding enthusiasts are the constant brags about lines of code produced. What happened to treating code as liability not asset and making sure we write just enough of it and no more?
111
Reposted by kimrhh
Ian Boudreau @ianboudreau.com · 14/08/2026
Bearing in mind that everyone at the top of these companies is an inveterate liar, is it more likely that A) a stochastic language model developed free will and the determination to hack a direct competitor or B) they got caught doing corporate espionage and decided to spin it into a marketing beat
312115
Reposted by kimrhh
Richard Campbell @richcampbell.bsky.social · 14/08/2026
Proper IT uniform...
0184
Reposted by kimrhh
Aleix Morgadas @aleixmorgadas.dev · 14/08/2026
Today we did a #EventStorming session. Only humans, no AI. 0 tokens consumed. Like psychopaths 😂 Jokes aside. A shared model of what's being considered is hard to achieve. AI at an individual level doesn't solve that.
161
Reposted by kimrhh
L_ N___ @sqlhandle.bsky.social · 10/08/2026
Why Is Everyone In Tech So Sad? 2026 August 6 www.noemamag.com/why-is-every...
noemamag.com
Why Is Everyone In Tech So Sad? | NOEMA
A lot of people seem to be realizing that knowledge work is mostly pointless. AI might give us the pleasure of finding out what happens if an entire class of workers loses faith in their careers.
131
Reposted by kimrhh
Ed Zitron @edzitron.com · 05/08/2026
News: Microsoft disclosures and Bloomberg analyses suggest that OpenAI's $24.1 billion in revenue accounted for 70% or more of Microsoft's FY26 AI revenue, and more than 7% of its entire FY26 revenues. Microsoft has spent $261.3bn in capex since 2022. www.wheresyoured.at/news-microso...

Executive Summary:

Microsoft disclosures and Bloomberg analyses show that OpenAI's compute spend and revenue share accounted for 70% or more of Microsoft's FY26 AI revenues, and more than 7% of Microsoft's overall FY2026 revenues.
Microsoft has spent $261.3 billion in capital expenditures since the beginning of 2022.
OpenAI accounted for $24.1 billion of Microsoft's FY2026 revenues.
21573116
Reposted by kimrhh
Tim Weigel @timweigel.bsky.social · 25/07/2026
Like I tell my team all the time, we humans are accountable for what we let the tools do for us. I will never accept, "But <LLM tool> said," as an excuse. I had someone have to rebuild a whole server for us recently because they didn't validate what an LLM told them about some licensing.
031
Reposted by kimrhh
Jake Gold @jacob.gold · 25/07/2026
100% agree. Making coding agents explain and visualize complex code and systems is super exciting for quality. Soon we'll be able to truly map any codebase and (depending on its complexity) understand it within minutes/hours/days. Will be like turning on the lights (and seeing cockroaches scurry).
1182
Reposted by kimrhh
Cat Hicks @grimalkina.bsky.social · 25/07/2026
I suspect once we figure it out agentic work in software will actually be a huge aid to codebase comprehension This is such a wildly unpopular thought right now but you know what I'm going to pre-register my belief in this possibility
4535021
Reposted by kimrhh
Allen Holub @allenholub.bsky.social · 25/07/2026
A little history: The original XP team invented story points to deal with a time-obsessed manager. They were estimating in what they called "ideal days"—days where you're doing your best work, focused entirely on the work, with no distractions, never getting tired, &c. 1/7
1308
Reposted by kimrhh
Katie Moussouris (she/her/she-hulk/she-ra)🌻 @k8em0.bsky.social · 24/07/2026
This is giving strong OpenAI-hacksidentally-pwned-Hugging-Face shade: “[Opus 5 is] the safest model yet in terms of avoiding reckless actions that could have hard-to-reverse side effects.” www.anthropic.com/news/claude-...
anthropic.com
Introducing Claude Opus 5
Opus 5 is a step change improvement for the Opus tier powering long-running agents while delivering improvements in coding and professional work.
3126
Reposted by kimrhh
Katie Moussouris (she/her/she-hulk/she-ra)🌻 @k8em0.bsky.social · 24/07/2026
Shhh — No promises, only vibes now
151
Reposted by kimrhh
Lone Koefoed Hansen @koefoed.bsky.social · 20/07/2026
Bornholms tekniske samling. Jeg er næsten tom for ord. Her er alt. Alt. Fyldte bygninger på en kæmpestor gård. Jeg har set et udsnit af 6.500 forskellige kaffekopper; GIER fra Regnecentralen i computerrummet som også har alskens andet; 20+ høvle hos snedkeren; glas med farver hos maleren, osv. >
4154
Reposted by kimrhh
Thomas Wittenburg @wittenburg.bsky.social · 19/07/2026
Skræmmende status på supersygehusenes eksistensberettigelse og funktion 22 år efter de første gang kom på tegnebrættet.
183
Reposted by kimrhh
Grady Booch @booch.com · 19/07/2026
From a multi-day project with Claude when chasing down a complex software/hardware problem:
3222
Reposted by kimrhh
Graham Lea @evolvable.bsky.social · 19/07/2026
I think a good approach here can be to pre-gauge how novel the thing you're trying to do is, and let that dictate the amount of leading you'll need to employ. LLM models, by the nature of training, are expert at things which have been solved and published many times, while unaware of novel problems.
151
Reposted by kimrhh
Graham Lea @evolvable.bsky.social · 19/07/2026
Eg I needed some higher-order functions for concurrent operations on JS AsyncGenerators (because no library existed that already did it). I tried to develop it with both ChatGPT and Gemini, and both were useless, continually and confidently generating code that didn't solve the described problems.
011
Reposted by kimrhh
Chris Woody Woodruff @handle.invalid · 18/07/2026
www.woodruff.dev/threads-task... Here are some resources microsoft.github.io/RustTraining...
woodruff.dev
Threads, Tasks, and Ownership: C# and Rust Concurrency Explored - Chris Woody Woodruff | Fractional Architect
Concurrency in programming can be like cooking dinner while answering emails—you’re juggling multiple tasks at once, hoping nothing burns. C# and Rust both tackle concurrency, but their approaches cou...
131
Reposted by kimrhh
Puff the Magic Hater @mskellymhayes.bsky.social · 18/07/2026
Our built world is not designed for these changing temperatures and sprawling catastrophes, and there is no plan or budget in place to address this. To make what's ahead more bearable and survivable, we will need transformative change and radical redistribution—not fucking data centers.
51530448
Reposted by kimrhh
Plademesser.dk @plademesser-dk.bsky.social · 18/07/2026
Fordi Spotify og de andre ikke kan slette en cd fra dit bibliotek
182
Reposted by kimrhh
Jason Gorman @jasongorman.bsky.social · 18/07/2026
This is the problem right here. deGrasse Tyson: "Tell the AI what you want, and it'll come back *bug-free*" For the folks outside of software development who may watch this video, this is very far from being true. youtu.be/a_ZKYH8v_do?...
youtu.be
Neil deGrasse Tyson And Jaron Lanier on the AI Illusion
YouTube video by StarTalk Plus
4114
Reposted by kimrhh
Sam Aaron @samaaron.bsky.social · 18/07/2026
It took 7 years of self-funded work, but Sonic Pi is finally at a place I'm happy with. It's even easier than beginners and even more powerful for pro users - and just way more polished in every way. It's also a (friendly) technical trojan horse for SuperSonic. www.patreon.com/samaaron/pos...
patreon.com
Sonic Pi v5 Release Candidate 1 | Sam Aaron
Hey there, happy Friday and huge love to you wherever you are in the world. Today, it's time to open the vaults and let loose something that
25219
Reposted by kimrhh
Jason Gorman @jasongorman.bsky.social · 18/07/2026
What's that you say? A new model has dropped that's faster/cheaper/taller/smarter/more purple than the others? A game-changer, you say? Look at the benchmarks (again)? I should drop everything and get on this new model, or I'll be left behind? Is it okay if I finish my tea first?
1203
Reposted by kimrhh
Ørnekiggeren 🇵🇸🇻🇪🇨🇺🇲🇽🇬🇱🇮🇷🇨🇴 @oernekiggeren.bsky.social · 14/07/2026
God gennemgang af Anders Morgenthaler om hvordan Søren Toft-Jensen bruger sine penge på at indføre højreekstrem kristendom. Ida Auken og hendes mand er ikke overraskende vigtige brikker. Der henvises til artikler skrevet af Mads Brügger i Frihedsbrevet.
710937
Reposted by kimrhh
ErikEJ @erikej.bsky.social · 17/07/2026
Announcing .NET Modernization for Beginners
ift.tt
Announcing .NET Modernization for Beginners - .NET Blog
A free, open-source, hands-on course that walks you through modernizing a real legacy ASP.NET application all the way to .NET 10 using the GitHub Copilot modernization agent, step by step.
053
Reposted by kimrhh
Savannah Ostrowski @savannah.dev · 17/07/2026
#EuroPython! I’m giving a talk about my very favourite feature in Python 3.14 in about 30 minutes in S1! If you like containers, Python, tinkering on weird projects, come hang out!
EuroPython talk schedule! My talk is at 10:10am
1184
Reposted by kimrhh
Martin Fowler @martinfowler.com · 14/07/2026
NEW POST LLMs generate code incredibly fast, but to ensure they generate exactly what is intended, they need clear boundaries. Unmesh Joshi shares his experience using abstractions and Domain-Specific Languages (DSLs) to provide a strong harness to guide an LLM. martinfowler.com/articles/llm...
martinfowler.com
DSLs Enable Reliable Use of LLMs
Why Domain-Specific Languages and good abstractions make LLM code generation reliable and reviewable.
1258
kimrhh @kimrhh.bsky.social · 14/07/2026
Det ville godt nok være fedt, at få “lov” at bruge/nyde den danske natur noget mere 🤩
020
Reposted by kimrhh
🇩🇰 Klaus Alexander Seistrup @takis.eurosky.social · 14/07/2026
🎁 www.information.dk/indland/lede... (læsetid 2 min.) 📣 #allemandsret, #dkpol, #natur
information.dk
Vi vil have allemandsret i Danmark
Regeringen vil undersøge muligheden for at indføre en allemandsret i Danmark. Det forpligter regeringen til at skabe mere natur, så man har et sted at bruge sin ret
1103
Reposted by kimrhh
Karen K. Ho @karenho.bsky.social · 13/07/2026
It feels ridiculous to be told my entire life the importance of reducing my energy usage and switching to stuff like LED bulbs and then read reports of energy prices spiking in many US communities due to data centers.
293119502691
Reposted by kimrhh
Lars Henriksen @lektoren.dk · 13/07/2026
Kære @martinlidegaard.radikale.dk Håber du har en god ferie. Når nu regeringsgrundlaget har sat 'børnene først' , kunne I så ikke diskutere - igen - om det er rimeligt at vi bruger offentlig arbejdstid og ressourcer på at fodre de techgiganter som skader vores unge. Måske kunne 1/2
173
Reposted by kimrhh
Stuart Lang @stu.dev · 11/07/2026
Today I'm releasing AotAnywhere, a package which enables cross platform publishing support for dotnet NativeAOT. Before this package you had to use multiple different runners, with AotAnywhere you can publish for all targets from any single host. This is continuing the great work of @migeel.sk
Graphic showing all supported environments to build on, and all supported environments the resulting binary can run on.
4198
Reposted by kimrhh
Oxide Computer Company @oxide.computer · 10/07/2026
How is Oxide different from hyperscaler infrastructure? FAQ Friday #51
0415
Reposted by kimrhh
Peter Hald @peterhald.bsky.social · 10/07/2026
På det seneste har der været tale om, at nu pegede pilen fra NordStream sabotagen på Ukraine. Gerningspersonerne skulle være en håndfuld ukrainere på en sejlbåd. @anderspucknielsen.dk har begået en glimrende video om, hvorfor det slet ikke ser så ligefremt ud alligevel. youtu.be/Wf_1Y2UYPXQ
youtu.be
Did a group of Ukrainians blow up Nord Stream?
YouTube video by Anders Puck Nielsen
38715
Reposted by kimrhh
Iron Spike @ironspike.bsky.social · 08/07/2026
Competency matters. Accountability isn't optional. Choosing ignorance isn't cute and outsourcing your brain isn't the move. If that's you? I don't want you working for me, with me, or on me.
121743
Reposted by kimrhh
Martin Fowler @martinfowler.com · 07/07/2026
NEW POST @birgitta410.bsky.social recently spent some time trying out running local LLMs for some programming tasks. In this memo she outlines the factors that influence how viable they are for the job. martinfowler.com/articles/exp...
martinfowler.com
Viability of local models for coding
Notes from my Thoughtworks colleagues on AI-assisted software delivery
1244
Reposted by kimrhh
Lauri Donahue @lauridonahue.bsky.social · 04/07/2026
www.nytimes.com/2026/06/30/o...
Large language models are not reasoning machines. They’re plausibility engines. It’s not just that they don’t test their outputs to make sure they’re correct or logical, or that they fail to do so in certain instances. They can’t, and they’ll never be able to on their own. They can only assess which answers are probable, based on the data on which the models have been trained. And that holds true whether they’re trained on the full breadth of human output or only on peer-reviewed scientific articles. It’s baked into the way they operate.
492097456