Stephen Dolan

“Kids I have good news. Daddy is out of Claude tokens until 3PM. He has time to play with you now.”

What a time to be alive. areyougoingexponential.vercel.app/stephendolan

@jorilallo @linear The agent has access to the MCP too - if you want a beta tester, I’m in!

@Stephen_Dolan @linear Probably first pass will be in MCP but as it builds on the API expect updates there as well

Our goal is to make Linear completely controllable through agent, in Slack, or inside Linear. Linear agent is soon just Linear and it will help control all your other agents.

@r00k @AmpCode This is a great fallback to add too, when you’re moving quickly: vercel.com/docs/og-image-…

When I need an og:image, I ask @AmpCode to generate 30 and "make me a tinder for them". Works great.

Is the agent autonomously managing everything in your life with no input from you in the room with us right now?

Too many people are concentrating on the one-shot workflow, and waiting for models to get better so they can one-shot more problems. The better workflow. which is already possible today, is to declaratively program and loop. You don't have to wait for models to improve.

Making devs finally write docs by calling them “skills” is a great arb

We've been developing a revolutionary take on this @tuple. You talk to a human being while you're making the thing ✨ Early testing is showing a lot of promise.

We're reshaping how marketing is done at Vercel. Today, everybody in marketing spent the afternoon building agents that help them automate the boring, tedious parts of their jobs. Every agent worked by the end. It was magical. We're in a new world. Everyone can cook.

Yet another time in my life where having the higher-order skill of GTD-based task management makes my road amidst change a bit easier!

AI workflows are technically impressive, but there’s a deeper reason people are really amped about AI agents. This isn’t just new tech, it’s new psychology. Until now, very few people have known what it feels like to delegate to total competency. If you manage great people, or lead great teams, you know how it feels to put someone in charge who will get it done, get it done right, and get it done without drama. That kind of delegation is pure joy. Delegating to competency lets you forget about it completely. That’s real leverage. And now anyone can experience that. Everyone can feel it. And it feels fucking great. That’s a big reason why the excitement is real, and fully justified.

We need a shorthand way of saying: "An AI did the work, but I vouch for the result" Saying "I did it" feels slightly sketchy, but saying "Claude did it" feels like avoiding responsibility

You can't let the slop and cringe deny you the wonder of AI. This is the most exciting thing we've made computers do since we connected them to the internet. If you spent 2025 being pessimistic or skeptical on AI, why not give the start of 2026 a try with optimism and curiosity?

We're rapidly entering a time where building is more fun than consuming.

the goal isn’t one-shotting, it’s rapid iteration

I still love to write code by hand, but you're cheating yourself if you don't at least have a look at what the frontier is like at the moment. This is an incredible time to be alive and to be into computers.

@adamwathan I basically don’t use any desktop clients for my non-code Tuple questions and research. Setting up memory, templates for specific kind of chats about revenue data or security questionnaires, etc all just feel more intuitive to me in my terminal.

Is anyone using Claude Code or similar for stuff that isn't coding at all, purely because it's nice to have access to local files and stuff? I use AI as a thinking partner constantly, and not having persistent references is starting to become annoying, but worried the CLI tools are tuned too much for programming and not business strategy for example.

I released an updated version of the Scry CLI that allows you to define templates that you can use for your ephemeral Claude code sessions: github.com/stephendolan/s… I was able to use it this morning to one-shot a security questionnaire (plus 10m to review for accuracy). Neat!