Had this convo with friends and workmates. I'm more excited than ever, same joy that's kept me hooked for 10 years is still there. It was never the code, it was solving the user's problem. AI lets me focus fully on that part and still ship clean solutions.
mwikala retweeted
It all starts with a Polycam scan 🙏
Thanks for sharing, Mwikala! 🤳
mwikala retweeted
Friendly reminder that if you have any MySQL 8.0 databases running on AWS still, you will be charged for RDS Extended Support this month and it's not cheap ($0.10 per vCPU-hour)
mwikala retweeted
STACKED PRS ARE LIVE ON GITHUB RIGHT NOW gh.io/stacks 🚀
i work with people across multiple timezones. i needed a little widget to compare timezones.
you can just do things.
quickshell + @OmarchyLinux quattro 🔥
i've also had this little plugin to handle auto light and dark mode themeing on omarchy for a while but quickshell makes adding a UI for configuring light mode and dark mode a lot easier.
i work with people across multiple timezones. i needed a little widget to compare timezones.
you can just do things.
quickshell + @OmarchyLinux quattro 🔥
1
3
286
.@alexandersix_ just wanted to say that I finally followed the advice from your Laracon Denver talk last year and I am officially a tmux user. I love it, thank you!
Also thank you to @dhh for making it so easy in Omarchy Quattro.
2
6
862
this is the way! I can never go back to just type prompting
im even voice prompting to my @NousResearch Hermes agent in discord
One pattern I find useful for working with LLMs is a nice long ramble session. Sometimes the LLM needs more bits to understand what you're trying to achieve, but you're too lazy to type them. In these cases I like to lean back, switch to /voice and just ramble for like 10 minutes, total mess, anything goes, full stream of consciousness. Sometimes I declare it up top, something like "switching to speech recognition sorry for any typos...". Sometimes I turn it into a small interview of a few turns. But I find that the LLMs are somehow very good at reconstructing long incoherent rambles and often their echo of your own tangle of thoughts comes out quite a bit cleaner than what you started with. The result is that you improve the mind meld and have to correct things less from that point on.
2
5
1,176
mwikala retweeted
One pattern I find useful for working with LLMs is a nice long ramble session. Sometimes the LLM needs more bits to understand what you're trying to achieve, but you're too lazy to type them. In these cases I like to lean back, switch to /voice and just ramble for like 10 minutes, total mess, anything goes, full stream of consciousness. Sometimes I declare it up top, something like "switching to speech recognition sorry for any typos...". Sometimes I turn it into a small interview of a few turns. But I find that the LLMs are somehow very good at reconstructing long incoherent rambles and often their echo of your own tangle of thoughts comes out quite a bit cleaner than what you started with. The result is that you improve the mind meld and have to correct things less from that point on.
Had this convo with friends and workmates. I'm more excited than ever, same joy that's kept me hooked for 10 years is still there. It was never the code, it was solving the user's problem. AI lets me focus fully on that part and still ship clean solutions.
1. OpenCode
2. Claude Code
3. Codex CLI
I'm running two subscriptions for this setup:
- $20 for GPT
- $20 for Claude
How many AI tools have you installed on your machine right now?
At least 8 for me right now, probably more that I not know of 😅
- @getpolyscope
- Solo
- Claude Code CLI
- Claude Native
- Codex CLI
- Codex Native
- @opencode
- @cursor
mwikala retweeted
Personal update: I've joined Anthropic. I think the next few years at the frontier of LLMs will be especially formative. I am very excited to join the team here and get back to R&D. I remain deeply passionate about education and plan to resume my work on it in time.

















