Lars Lykke
@llykke.bsky.social
Agent of change and father of 3 in a world of software development
Dependabot noise is optional. Group routine version updates, slow their cadence, and keep security fixes fast. Here’s how: github.blog/security/sup...
Tame Dependabot: Group your updates, slow the cadence, keep security fast
Here's how grouping updates, slowing the cadence, and keeping security fixes fast cut the noise on a Microsoft open source project.
github.blog
New article: Does code quality still matter? blog.ploeh.dk/2026/07/23/d...
Does code quality still matter?
I honestly don't know, but I guess we'll find out.
blog.ploeh.dk
I’m starting to see a trend where agents skills are being used as an alternative to fixing existing developer experience problems and it’s starting to annoy me. Skills are making people lazy and are becoming a band aid, abstracting away the bad experience and leaving the agents to work it all out.
Architect Elevator Complete Collection by Gregor Hohpe is the featured bundle of ebooks 📚 on Leanpub! Own all three books in the Architect Elevator series for 900 pages of large-scale IT architecture! … leanpub.com/b/architecte...
The thing about apps that take meeting notes is that even if they were perfect, I personally wouldn't use them, because I cannot focus on meetings but I can focus on writing, so by writing out the meeting I trick myself into paying attention
Good decisions lose value if nobody remembers the reasoning behind them. Social Decision Records and Team APIs help teams capture the thinking behind decisions, making collaboration much easier over time. Take a look →
Social Decision Records and Team APIs: Documenting Human Interactions for Better Collaboration — Team Topologies - Organizing for fast flow of value
SDRs are an easy, lightweight tool to help you analyze, negotiate, and formally record the decisions of a complex sociotechnical system. They are highly flexible and can be adapted to fit your team's specific people management or delivery concerns. When combined with the concept of a Team API , SD
teamtopologies.com
'Tis the last day to preorder The Psychology of Software Teams! By ordering today you'll join the POST advance squad which comes with benefits like a) I remember every early post forever b) bragging rights c) enormous increases to your persuasion d) a faint glowing aura www.drcathicks.com#book
Dr. Cat Hicks
Leading the science of software success
drcathicks.com
Cat Hicks@grimalkina.bsky.social · 4w agoSo how do you help the authors you want to help? - Preorder - Tell one specific person "you'd like this book" — this converts at a far higher rate than anything! - If you're on a team or in a reading group, suggest it there - Request it at your local library! www.drcathicks.com#book
Keep one person on an AI safety team who never uses LLMs, ever? The worry is a slow con, these tools quietly earning your trust. You'd want one person whose immunity stays intact.
The GitHub Copilot CLI for Beginners is the best starting point after the documentation: github.com/github/copil... I'd appreciate more insights into what one expects in terms of educational material specific to one programming language.
GitHub - github/copilot-cli-for-beginners: Learn how to get started using the GitHub Copilot CLI!
Learn how to get started using the GitHub Copilot CLI! - github/copilot-cli-for-beginners
github.com
A Wednesday lunch meeting turned "that's not possible" into "we shipped it last week". That is the short version of a Community of Practice that worked. A thread on the format that moves knowledge between teams the org chart was designed to keep apart.
alright friends and computationists, what are the signals you pick up on that make you think a team has high engineering discipline? (whatever "high" means to you) i'm not talking science or evidence. i'm talking casual conversation, heuristics, no evidence whatsoever, just radar. what trips it?
Tracking the "what's announced at Build" news for Windows developers. learn.microsoft.com/windows/apps...
What's new for Windows app developers - Windows apps
Learn what's new for developers in Windows 11 and tools
learn.microsoft.com
I started an article on providing sensors about the code's maintainability to coding agents. Part 1 and 2 out so far, about static code analysis, coupling, and modularity review martinfowler.com/articles/sen...
martinfowler.com
2 questions every team using Argo CD should ask about secrets management: Should changing a secret require an Argo CD sync? Should your secrets solution be coupled to Argo CD? The answer to both: no. Join us tomorrow on Argo Unpacked to hear more: www.linkedin.com/events/74540...
LinkedIn Login, Sign in | LinkedIn
linkedin.com
I'm not saying AI tooling can't help. But if we're not measuring actual outcomes, we're just confirming our own bias. Start measuring. Compare cycle times, defect rates, rework. Don't let a feeling of speed substitute for evidence of speed. >>>
When I was a corporate employee, my calendar was always packed. Meeting after meeting. By the end of the day, I'd been going 100 mph and gotten further behind. The organization rewarded people who were always slammed. But busy and productive are not the same thing. #TimeToThink #SelfLeadership
Most organisations don’t lack effort. They lack clarity on where to start. The Team Topologies assessment helps surface where things feel heavier than they should, and where to begin. Take the assessment here - https://survey.teamtopologies.com/tt-assessment
I feel this post from @johncutle.fish is about to become very important (for me) in terms of how work gets structured in my context… cutlefish.substack.com/p/tbm-414-le...
TBM 414: Legibility and Legitimacy
There is a big difference between legibility and legitimacy.
cutlefish.substack.com
"Being the Human in the Loop" - @kevlin.bsky.social. Typing isn't the bottleneck. Be the driver, not the road. Watch youtu.be/vpYJMr1pJRY
Being the Human in the Loop – Kevlin Henney
YouTube video by Agile meets Architecture
youtu.be
New article where I offer definitions & a mental model how to think about harness engineering as coding agent users. Building blocks at our disposal, dimensions and goals to consider; emerging practices, open questions; and of course, what role do humans play martinfowler.com/articles/har...
Harness engineering for coding agent users
A mental model for building trust in coding agents through feedforward guides, feedback sensors, and iterative harness engineering.
martinfowler.com
This post by @johncutle.fish really resonates with me cutlefish.substack.com/p/tbm-412-in... I don’t see any change in how work is performed - just people working faster - because that’s what expected with AI right? That behavior gets promoted and internalized. We need to work different - not more
TBM 412: Institutionalized Overload (Now With AI)
(I’ll be playing a bit of TBM catchup over the next couple days.
cutlefish.substack.com
This is one of the things collaborative software design is fundamentally about: making implicit knowledge explicit. The opinions, the preferences, the emotions behind a technical stance, they don't disappear when you ignore them. They just go underground and shape decisions anyway, invisibly. >>>
"AI Augmentation for Technical Leaders" - Birgitta Boeckeler. Can AI help with architectural decisions? Watch:& Subscribe. youtu.be/qB7rsbDfmQg
What AI augmentation means for technical leaders – Birgitta Boeckeler
YouTube video by Agile meets Architecture
youtu.be
🎉 That's a wrap with the last sessions from @andrewhl.bsky.social @bitboss.bsky.social and @josvanschouten.bsky.social 2 days, 28+ sessions, countless connections. From AI to Team Topologies to real-world solutions. Thank you. See you next year
And so it’s a wrap! I’ve been attending the Agile Meets Architecture 2026 #amaberlin @amaconf.bsky.social in Berlin and my head is still buzzing from all the insights on how we (could?) shape both systems, organisations and ways of working through architecture. …and sometimes the other way round 😉
If using agentic ai corresponds to ingesting more work into a system (e.g.generating code) but we’re not moving faster? So using ai didn’t remove the right bottleneck in a theory of constraints type of way? During a discussion I got the feeling that value-stream-mapping and LEAN are still relevant.
