Don Is Typing ⌨️

@AnthropicAI Alternatives I use: @ollama for local models @asklumo for private inference off device

@AnthropicAI affordable robust local inference can’t come fast enough 😐 👇👇👇

How do people seek guidance from Claude? We looked at 1M conversations to understand what questions people ask, how Claude responds, and where it slips into sycophancy. We used what we found to improve how we trained Opus 4.7 and Mythos Preview. anthropic.com/research/claud…

@itsolelehmann a thousand times no

anthropic's in-house philosopher thinks claude gets anxious. and when you trigger its anxiety, your outputs get worse. her name is amanda askell. she specializes in claude's psychology (how the model behaves, how it thinks about its own situation, what values it holds) in a recent interview she broke down how she thinks about prompting to pull the best out of claude. her core point: *how* you talk to claude affects its work just as much as *what* you say. newer claude models suffer from what she calls "criticism spirals" they expect you'll come in harsh, so they default to playing it safe. when the model is spending its energy on self-protection, the actual work suffers. output comes out hedgier, more apologetic, blander, and the worst of all: overly agreeable (even when you're wrong). the reason why comes down to training data: every new model is trained on internet discourse about previous models. and a lot of that discourse is negative: > rants about token limits > complaints when it messes up > people calling it nerfed the next model absorbs all of that. it starts expecting you to be harsh before you've typed a word the same thing plays out in your own session, in real time. every message you send is data the model reads to figure out what kind of person it's dealing with. open cold and hostile, and it braces. open clean and direct, and it relaxes into the work. when you open a session with threats ("don't hallucinate, this is critical, don't mess this up")... you prime the model for defensive mode before it even sees the task defensive mode produces the exact output you don't want: cautious, over-qualified, and refusing to take a real swing so here's the actionable playbook for putting claude in a "good mood" (so you get optimal outputs): 1. use positive framing. "write in short punchy sentences" beats "don't write long sentences." positive instructions give the model a clear target to hit. strings of "don't do this, don't do that" push it into paranoid over-checking where every token goes toward avoiding failure modes 2. give it explicit permission to disagree. drop a line like "push back if you see a better angle" or "tell me if i'm asking for the wrong thing." without this, claude defaults to agreeable compliance (which is the enemy of good creative work) 3. open with respect. if your first message is "are you seriously going to get this wrong again?" you've set the tone for the entire session. if you need to flag something, frame it as a clean instruction for this session. skip the running complaint 4. when claude messes up, don't reprimand it. insults, "you stupid bot" energy, hostile swearing aimed at the model, all of it reinforces the anxious mode you're trying to avoid. 5. kill apology spirals fast. when claude starts over-apologizing ("you're right, i should have been more careful, let me try harder") cut it off. say "all good, here's what i want next." letting the spiral run reinforces the anxious mode for every response that follows 6. ask for opinions alongside execution. "what would you do here?" "what's missing?" "where do you see friction?" these questions assume competence and pull richer output than pure task prompts 7. in long sessions, refresh the frame. if a conversation has been heavy on correction, claude gets increasingly cautious. every so often reset: "this is great, keep going." feels weird to tell an ai it's doing well but it measurably shifts the next 10 responses your prompts are the working environment you're creating for the model tone, trust, permission to take a position, the absence of threats... claude picks up on all of it. so take care of the model, and it'll take care of the work.

I was blessed with a free claude code promo someone gifted me a few weeks ago. Today sadly it has expired. I'm really shocked at how much I used it.

Go give this guy a follow and some 👀

@OpenAI is going to have to offer something more "on the rails" than @openclaw if they want to compete with @claudeai's cowork. Yes, we've had it with the rate limit nonsense but there has to be a clean and trusty alternative.

One day we'll be all done with this paid token nonsense. I'm already seeing what look like viable local alternatives. Would 100% drop a few grand in order to be able to do inference locally for free.

This is the way

Gardening is highly recommended btw, especially for folks who tap on keyboards all day

Mint tea courtesy of the garden 🌱

@JakeATech Nice! I haven't done anything special with the actual Jellyfin setup. It's pretty good out of the box IMO. I use it for video (via firestick app mostly) and for music. For music I found two apps to be the best Finer and Manet monk-studio.com/finer tilosoftware.io/manet/

@donistyping I have a Teramaster I run my Plex on 😅 I'd like to give jellyfin a try again though. Any particular tips you found useful for it? Or is the standard setup good enough?

For self-hosted media streaming, what do you use? I currently just use Plex in a container, but want to know what else people use!

@RileyRalmuto It’s good, definitely try it. I can’t seem to get it to bypass all permissions to run scripts but it’s early on. Definitely worth the switch from OC.

I still haven't played with Claude Dispatch. anyone else explored it yet?

@jrswab lol not sure about that. I was considering adapting some kind of e-ink alternative for my main driver monitor which is where my terminal/ide usually is. My millennial eyes are going through a midlife crisis I’m 100% sure light mode would exacerbate.

100x devs use light mode.

Downloaded Antigravity to give it another go - AMA It's on free mode rn so I know they're watching me but it's probably worth it.

@sundeep @steipete We’ve crossed over into a place where now it’s so obvious these companies are trying to just make money.

“If your $500K engineer isn’t burning at least $250K in tokens, something is wrong.”

@MatthewBerman I just tried it today after getting frustrated my openclaw fell over while I was trying to install a new model. It is good. Better than what I was getting from OC.

* They've BUILT OpenClaw

What a horrible doppelgänger to have lol