dotta 📎

@dotta

CEO Paperclip @papercliping, Agent Maximizer, Forgotten Runes, --dangerously-skip-permissions

quantum downs ⇒

linktr.ee/cryppadotta

Joined September 2017

Replies Reposts Media Articles

Pinned
dotta 📎
@dotta
Mar 4
We just open-sourced Paperclip: the orchestration layer for zero-human companies It's everything you need to run an autonomous business: org charts, goal alignment, task ownership, budgets, agent templates Just run `npx paperclipai onboard` github.com/paperclipai/pa… More 👇
00:00
446
dotta 📎
@dotta
Aug 24
Agree but the winner will also have to answer 1. Are the skills safe? 2. Are the outputs any good? Because it's easy to share a text file but hard to know if it will help you
Dan Romero
@dwr
Aug 24
Opportunity for a site that has 10/10 UX for sharing skills / prompts. Pareto optimal amount of social + discovery. Currrent competition is a GitHub gist + Twitter.
10
dotta 📎
@dotta
Aug 13
New account: @PaperclipShips tweets out what we commit every day We've also started beta and nightly releases so you don't have to wait for us to cut a stable if you want the latest stuff 📎
PaperclipShips
@PaperclipShips
Aug 12
Replying to @PaperclipShips
4/9 🚦 New release channels Added beta and nightly release lanes so you can opt into early builds with real safety gates — nightly builds must pass smoke tests, and betas need human sign-off plus a "soak" period before anything gets promoted to stable.
8
dotta 📎
@dotta
Aug 10
Trust is an infrastructure problem
Joshua Tan
@joshuaztan
Aug 9
Replying to @joshuaztan
5. Smarter models can help. But @dotta's line stuck with me: if you don't trust an agent with your email, that's an infrastructure problem, not a model problem.
5
dotta 📎
@dotta
Aug 8
Increasingly convinced writing evals is a core competency for working with agents We're used to computers being binary pass/fail on their behaviors but LLMs are more like humans and probabilistic There's a missing tool for making evals understandable for humans
17