vist dev

Secure founder pricing now, before it runs out → up with my thoughts →

[vist dev @vistdev]

@mattpocockuk Knew I should have gone!

[vist dev @vistdev]

@3chospirits @trq212 Same habit here! A living plan file as MD beats hoping the chat history stays coherent.
It gets saved automatically & is next to my tasks, so the next session (and me, the next day, because my own context switching can be suboptimal too!) can pick up the "why" and "how" instantly

[vist dev @vistdev]

@jordancodeadv @microDesignn You can use the superpowers skill set - SDD (subagent driven development) does this for you. Or you can cobble together your own variant that does this just the way you like.

[vist dev @vistdev]

@zak1387 You're probably not the problem. Models don't keep a durable project brain across sessions unless you give them one (notes/tasks the agent can re-read). Otherwise every fresh thread starts from zero, no matter how good the model is that day...
Don't rely on OpenClaw memory only?

[vist dev @vistdev]

If you're trying MCP for the first time: start with one project of markdown notes + tasks, not a giant memory dump.
Set up the connection and ask the AI "what's open and what did we decide last time?", and see if it actually answers...

https://t.co/Ou3UcJ7Pyg

[vist dev @vistdev]

@Zallordian @danellisona 🥳 still haven't hit rock bottom, yay!

[vist dev @vistdev]

@bcherny @jpreagan Medium has been my go to for most of my work.
Not everything requires superstar coding intelligence.

[vist dev @vistdev]

@Zallordian @danellisona Haha, I never tried that. I tend to stay away from the AI slop/bot posts… but maybe that’s not the right approach.

[Thariq @trq212]

the right way to use model capabilities is not to ship 10x more features to prod

it's to spend more time understanding your users, trying experiments, building prototypes, learning about things you don't understand so that you can ship things that actually work

[vist dev @vistdev]

@danellisona @Zallordian Not always 🫣… sometimes the conversation turns out to be with a marketing bot. 😞

[vist dev @vistdev]

@philaraujoPM Great observations! The solo company-brain attempt usually dies on maintenance (like most things, right?) ... so until the AI can read/write the notes without you babysitting every paste, it's doomed!
Are you making any difference between notes and decisions and so on?

[vist dev @vistdev]

@vinsonleow @grok @claudeai @AnthropicAI @ChatGPT @OpenAI Makes you wonder what they are _not_ shipping, but have been cooking internally...

[vist dev @vistdev]

@vinsonleow @claudeai @grok @ChatGPT @GeminiApp @Kimi_Moonshot Or: use Vist, with zero setup beyond a simple remote MCP connector, and get it all out of the box, no extra setup or tokens needed ;-)

[vist dev @vistdev]

This I am also very much looking forward to. Opus is still my go to for making plans…

[Claude @claudeai]

Opus 5.5 requires less compute to serve than Opus 5, and its pricing reflects that.

Our tests show that at default settings it will cost 40% less than Opus 5 on typical workloads.

[vist dev @vistdev]

I do hope the first reactions hold up and it’s more concise and down to earth than 5…

[Claude @claudeai]

Introducing Claude Opus 5.5, the first model in our new Claude 5.5 family.

It performs at the level of Claude Fable 5.1 for most tasks, and costs 40% less to run than Opus 5.

[vist dev @vistdev]

You open a fresh chat and re-explain your project, what not to do, and what you were mid-way through on Friday.

If that sounds familiar - it's not a model problem. You need a shared brain for you & the AI.

MCP, markdown, EU-hosted, simple yet efficient.
https://t.co/Ou3UcJ7hII

[claire vo 🖤 @clairevo]

so i have 40+ agents who go through my various emails, crms, slacks, etc and finds all the things i need to do.
they tell me to do it, and i ignore it.
so i vibe coded an app that gives all the agents a central place to tell me all the things to do.
i look at it, and i ignore it.
so then i open the app in codex and i tell codex to bully me into doing it...

[vist dev @vistdev]

@thedavidqu For me, I found that because I like to drive work from my phone as well, I needed something that is at the same time simpler (less setup, hosted) and more advanced (memory features) than Obsidian.

[vist dev @vistdev]

@kodevbuilds that really is the end goal, yes... one thing that I can run from any LLM I choose to work with, and gives the LLM all the knowledge and tools to help me run all my work

[vist dev @vistdev]

@mignano @ujjalcal That’s my feeling as well, and I’ll need to experiment with it asap. The pieces are there for us to combine them in new ways.