Published
All published posts
Is Compaction The Issue
I saw today in work chat something along the lines of “we need bigger context windows” “compaction times are holding us back”. Maybe I’m just blessed with the lack of lord jira, maybe juggle too many projects at once and they are all pretty much done when I get back. Maybe I do more long running specs and spend time making good plans that it does not matter.
Anyways the point I’m getting to is that if you think that compaction is your main issue slowing you down, and 10x this if you are a manager thinking this is what is slowing down your team you need to look at your workflow. Not because it sucks. Not only because it could be better. Because you are signing yourself and your team up for burnout if you are sitting there watching these things run like waiting for paint to dry and firing more prompts at them as soon as they are done. It feels easy. It feels like you are going fast. Its eating more brainpower than you think, and its not getting you to your...
Lets Land The Plane
Part of @steveyegge ’s gastown/beads is a prompt “Lets land the plane”. It’s very straightforward forward and what any sane human would probably do before finishing work, except the last part. The “generate a handoff prompt for the next session” was not something I’ve put much thought into. But now that I juggle 6 sessions at a time and often end up with 20 sessions open because I don’t want to close them and loose the last bit of context. This is what I need to keep from crippling my laptop memory from all of these stale sessions hanging around.
Taken from https://ianbull.com/posts/beads
I’ve been having issue with my keyboard disconnecting to my main desktop for awhile. Today I got a cheap bluetooh dongle in and am giving it a run this week to see how things go. The first step was to move it to the new adapter. I’ve never had multiple adapters installed so this was a new to me process.
I was able to do it all with the same keyboard, It did require some juggling between usb and bluetooth modes pluging and unplugging, two keyboards would be simpler to reason about.
I can’t be bothered to change my brain to think about this machine on a different zmk profile it is of absolute importance for it to remain on the same profile, otherwise this would be a simple bind to another empty profile.
Why not use a cable on desktop?
I dont mind cable, and have used one on this setup for years, but I have actually been picking up and moving this keyboard and using it with different devices.
I’ve got a big battery and performace cranked up, unless my machine is under load I do not notice any key lag.
I did it with bluetoothctl, I’m sure it could have been done with a gui like
blueberry or blueman.
bluetoothctl
# list adapters
list
select <old-adapter>
devices
# fin the MAC address of the device 42BLOCK
remove <42BLOCK_MAC>
Now I plugged into usb. And importantly cleared out the zmk profile. If you do not clear the profile your board does not go into pairing mode.
bluetoothctl
# switch adapters
select <new-adapter>
power on
agent on
default-agent
scan on
Put 42Block in pairing mode, then:
pair <42BLOCK_MAC>
trust <42BLOCK_MAC>
At this pint I saw this show up in the logs, I think there was some masking issues or something in zmk, output kept going out usb no matter what so I disconnected the keyboard and typed the passkey in, and it worked.
[agent] Passkey: 540044
Boom, it just started working right away.
bluetoothctl
connect <42BLOCK_MAC>
info <42BLOCK_MAC>
scan off
exit
I don’t want someone else running my agents
I don’t want to review the pr, I dont want to fight the mass of changes clobbered across the codebase. I want to own my platform. With everything changing with agents writing more code than I can imagine in a day work looks different now. I still want to work with real people. I want to collaborate on ideas. I want someone to bounce ideas off with. I want someone else in the war room with me on launch day, or when the whole thing goes down. But I don’t them slopping in my sandbox, if someone is going to be stirring the slop in my product I want it to be me. Work is feeling different now. New lines need to be drawn in new directions. Expectations are changing, the way work is completed is changing, and we are all here trying to figure out what this looks like moving forward.
Ping 38
When agents do the work its harder to recognize a dead end.
Learning to agent
All we are hearing lately is Agents are the future, something flipped around NOV 2025 with opus 4.5. It turned snake oil into action. It changed programmers will be replaced in 6 months to now. Not all of them, but probably most of us who are not extraordinary. If you fall into the camp of folks not adopting, I got no issues with that. No one is twisting your arm, well maybe your boss or cto is, thats on them. I don’t mean to say this is the future as in, get in or get left behind. I mean it as this is where your other engineers probably are, the junior to mid level engineers are here. If you are not trying to meet them where they are how are you going to lead them.
Studio Ghibli Images in the Wild
I just stumbled into an image in my org chart of someone who clearly turned themself into a Studio Ghibli character in chatgpt during the small window of time that it seemed to do this for everything. Its clearly the aesthetic that It would do by default for that week, then would not do it whatsoever. I’d link it, but its from an org chart. I mostly found it interesting how we now have these recognizable artifacts from specific moments in time.
Ping 36
I feel like there’s an inevitable phase to every ai/agentic worked feature/epic where you have to get in and chat with it 2025 style (except it actually works and doesn’t turn your project to shit). Planning is great, planning out epics for full orchestrator’s to churn for hours on is amazing, but it always leaves me with a handful of thorns multiplied by complexity level of things that I can shout a list of 6 items at a time that it can one shot. I haven’t seen anyone put a name to this phase yet, so I’m going to call it the UAT phase for now and it seems like a very necessary part of the SDLC. It was important before, but feels more so now as engineers distance themselves from the implementation.
Research, Plan, Implement
I heard this term yesterday, and I think a lot of people are missing out on step 1. It’s important to experiment with agents and learn what they can do well and what they cant, this changes every couple of weeks at this point. You might be spending hours planning something that could have been implemented right away, or maybe wasted time planning something that needed more research, more context engineering. Agents start fresh every session, they cant remember what you asked them to do 5 minutes ago in the other session, getting the right tokens in session is critical.
Today I learned that docker creates an empty /.dockerenv file to indicate that
you are running in a docker container. Other runtimes like podman commonly use
/run/.containerenv. kubernetes uses neither of these, the most common way to
detect if you are running in kubernetes is to check for the presence of the
KUBERNETES_SERVICE_HOST environment variable. There will also be a directory
at /var/run/secrets/kubernetes.io/serviceaccount that contains the service
account credentials if you are running in kubernetes.
Context Poisoning Was There All Along
I wrote some code by hand on Sunday. Sat down with my son and started building out a game in pygame from scratch. We went to google, we searched how to do something, we copy and pasted from the docs. Not because we are dumb, but because we cant remember some aspects of the pygame api. Now that these patterns are established we no longer have to google them, we simply grep our codebase and replicate the pattern. Easy right? It’s funny that it took ai to coin the term context poisoning even though it was there all along.
Calendar data could not load. Try again.