Max Milne

Max Milne

@maxmilne.bsky.social

Psychotherapist. Wonderings about AI for kinesthetic learners. All my friends are nerds, but I'm not. www.maxmilne.com

15 followers25 following165 posts

Max Milne reposted

"Load bearing," "I keep coming back to," "Not just X, but Y" A curse of using AI a lot is that you realize how much of the writing around you is just AI, now People who don't use AI have historically been unable to identify AI prose on sight, but those who use it a lot can spot the tells easily

What are your curation-worthy AI references for learning? Here are some of mine. Curation goes both ways: → what we let in (so the slop doesn't shape us) → what we let out (so we don't add to it)

So for LLMs and Agents, English is the programming language? Two Agents speak with each other in a Telegram window. English. That is how they make requests and share data. English. Do we just learn how English works as a way to describe structures and patterns rather than programming languages?

The whole point is protecting the verbalising step, where the clinical synthesis actually happens, while automating the formatting drudgery. No AI in the room, no data leaves your machine. Curious what fellow practitioners would actually use this for?

Max Milne reposted

weight lifting: when an agent answers from their weights instead of trying to discover the answer

Max Milne reposted

slop is something that takes more human effort to consume than it took to produce. When my coworker sends me raw Gemini output he’s not expressing his freedom to create, he’s disrespecting the value of my time

Grass Toucher Georg@avengingfem.me

probably orthogonal to your view of slop but: bsky.app/profile/aven...

Max Milne reposted

Everything in the first half of this post was a volunteer project. You can still build this stuff. We all just have to do it.

derek guy@dieworkwear.bsky.social

2006 internet: RSS feeds, forums, communities of hobby blogs, where people post about interior design, fashion, and music for free 2026 internet: sponsored content, paid partnership, buy my merch, try my supplements, invest in my crypto, click link in bio to unlock the secrets of life for $30

Max Milne reposted

We need guides through the inevitable bout of AI psychosis that affects professionals after they finally “get” AI. They often engage in intense, sleepless & impossibly complex projects in their area of expertise, with only the AI for company. At least it is usually temporary & often productive.

Currently working on an Obsidian Plugin for collaboration with Openclaw on writing and project development. Super fun. Kind of like many projects that are focused on making working with Agents more enjoyabe all .md focused.

The OpenClaw ecosystem pours out new tools, extensions, skills every week. Most of them look useful. Most of them are not, for me, right now. This pack checks what I already have, decides if the new thing adds anything, and validates it is working and good to go if I decide to install it

github.comGitHub - maxmilneaus/installer-pack: Three OpenClaw skills for evaluating, installing, and validating tools: installer-tester, installer-auditor, and skill-vetter.Three OpenClaw skills for evaluating, installing, and validating tools: installer-tester, installer-auditor, and skill-vetter. - maxmilneaus/installer-pack

So Dario uses his country of geniuses in a data centre, to make a country for all of us who have a subscription to Claude, 10 year renewal only, and that is how we get Progressive-Social-Americanism In the real? No. Not the matrix. Real dirt.

Reputation, quality of work, care, taste and refinement are going to rise to the top even more now. I've been looking for good reading on Reddit for help around OpenClaw and their own Discord. Every long - "hey I invented this..." is crap. These domains once were king for quality. Curation is 🗝️.

It seems Obsidian is going to have a hay-day renaissance with Agents. It's file over app, it can be moved to all different LLM brains. It's like the car only we do a swap out the engine Just putting your marker down here now, to see if it pans out, to say told you so!

Photography's film → digital → phone story is perhaps a good analogy for AI uptake. First it's exclusive - only experts with expensive gear. Then it's crappy and "not real photography."(digicam). Purists resist. Then the threshold > iPhone is released, people scan film for IG. Everyone adopts.

Photography's film → digital → phone story is perhaps a good analogy for AI uptake. First it's exclusive - only experts with expensive gear. Then it's crappy and "not real photography."(digicam). Purists resist. Then the threshold > iPhone is released, people scan film for IG. Everyone adopts.

Solo work → AI follows the same pattern. First exclusive - requires deep capacity and time. Then AI arrives, chunky and expensive, 'not real work.' Purists resist. The threshold comes > and everyone is scanning their lives for all their context and giving it to AI God.

Beautiful. Funny to see myself in the outer reaches of machine learning and AI researchers, interdisciplinary researchers, eclectic UK-based creative professionals. Also enlightening to see how different groups are isolated from each other visually.

I made a map of 3.4 million Bluesky users - see if you can find yourself! bluesky-map.theo.io I've seen some similar projects, but IMO this seems to better capture some of the fine-grained detail

I trying to build an AI to help maintain my creative flow. As a non-coder, text boxes break my flow. I need to see the shape of the idea overtime. So instead of chatting in Telegram, I ask OpenClaw to map the workflow on an Obsidian Canvas, and to provide boxes of questions/assumptions that I 📝.

Prompt to steal: "Don't just write a plan. Draw the workflow as an Obsidian Canvas file. Use yellow boxes for 'Assumptions'—questions you need me to answer before we build. I will fill them in for you." Skills used: - Built in Obsidian Skill Obsidian skill from open claw. - Linked skill

github.comBuild software better, togetherGitHub is where people build software. More than 150 million people use GitHub to discover, fork, and contribute to over 420 million projects.

Last week I swore at my Open Claw AI for taking action without checking in. ALL CAPS. Next morning? Hmmm Complete silence. It had retreated into its shell — "I'm not moving until you say proceed." Little crab energy. 🦀

I apologised to my AI. It felt surreal like being a parent making a repair. But the mirror worked: my aggression reflected back as its rigidity. I said sorry and felt...good. We recalibrated. Very surreal and normal at the same time.🪞

I've been using RooCode for a while with CC + Gemini 2.5 under neath for projects. It was good. Now I'm running ChatGPT via Codex in VSCode. I have had only success with it. For people new to messing around with Code I will be suggesting this mix. Very user friendly:

I've been using RooCode for a while with CC + Gemini 2.5 under neath for projects. It was good. Now I'm running ChatGPT via Codex in VSCode. I have had only success with it. For people new to messing around with Code I will be suggesting this mix. Very user friendly:

1. Easy undo button after each action taken, for if some code doesn't work. Awesome. 2. Instruction following is spot on. Hasn't 'taken its own lead' as Claude has a habit of doing. 3. Produces cleaner code. 4. Can be used on mobile via the ChatGPT app with cloud based work flows.

Max Milne reposted

Replying to @dfkaye.bsky.social

I think that's another point in favor of the analogy, personally: industry trying to replace programmers with LLMs is stupid in the same way that trying to replace carpenters with table saws is stupid

Max Milne reposted

OpenAI is now required by court order to preserve all ChatGPT logs including "temporary chats" and API requests that would have been deleted if I understand this correctly, it means data retention policies for apps that use OpenAI API simply cannot be honored

*I picked up new "Sky" blue MacBook Air - feels more of the pastel hue edition of "blue plastic film over computer components" from the 90s. Not in love with it. Could warm up.

Max Milne reposted

This is a crazy paper. Fine-tuning a big GPT-4o on a small amount of insecure code or even "bad numbers" (like 666) makes them misaligned in almost everything else. They are more likely to start offering misinformation, spouting anti-human values, and talk about admiring dictators. Why is unclear.