AI agents don't come with a manual either.
Every AI tool ships with a demo video. I'm building Restless to work out what it actually takes for people and agents to get real work done together, and this is where I write it down: what shipped, what broke, how Restless stacks up against the other tools, and plain-English explainers for anyone who's been told to "just use agents" and isn't sure what that means. Expect me to be wrong in public sometimes, and to work out why.
- →
Focus on your craft. Save the grunt work for AI.
Today we're launching Restless: a place where you and your team, human and AI, run a business together.
- →
How Restless ships, and what "development preview" means.
Software that runs your company shouldn't break your company when it updates. The promises I make about releases, and the ones I don't yet.
- →
Every handoff I've seen fail broke in one of three ways.
Watching agents pass work to each other, the failures stopped looking random. They're the same three mistakes, over and over, and people make them too.
- →
The process that wouldn't die, and what it taught me about cleanup.
An agent started a game, forgot about it, and the game quietly ate half a machine for most of a day. Killing it didn't work. Here's why, and the rule I live by now.
- →
What a good AI critic actually catches.
When I added a reviewer agent, the interesting part wasn't whether it found problems. It was which problems, and why a reviewer that sees too much stops seeing anything.
- →
Chat, workflow, agent or team? Picking the right shape for the job.
Most AI disappointment comes from using the wrong shape of tool, not the wrong model. Here's a simple way to pick.
- →
What's an AI agent, really?
Less magic than the demos, more useful than the sceptics think. A plain-English guide for anyone who's been told to "just use agents".
- →
How to brief an AI (or a person) so the work comes back right.
Most bad AI output starts with a bad handoff. Five things every brief needs, and three it shouldn't have.
- →
Can you undo it? The one question before letting AI act.
Approving everything burns you out. Approving nothing is reckless. Here's a simple way to decide how much freedom an AI agent gets.
- →
Where the money really goes when AI does the work.
Prices change every month. What drives the bill doesn't. A plain guide to what makes AI work expensive, and the second budget most people forget.
- →
"Done" is a claim, not a fact. How to check AI work.
AI will happily tell you it finished, the tests pass and everything looks great. Here's how to check without redoing the work yourself.
- →
Writing your company's voice so an AI can actually follow it.
"Friendly but professional" means nothing to a model. How to write a voice guide that works, and how to notice when the output drifts.
- →
Why your AI forgets, and what "memory" really means.
Context windows, chat memory and a company's records are three very different things. Knowing which is which explains most of the forgetting.
- →
Own the computer your company runs on.
Your laptop is too risky, a throwaway sandbox is too forgetful. Why AI agents need a computer of their own, and what self-hosting actually gets you.
- →
Restless vs one big chat: when one chat is honestly fine.
ChatGPT, Claude and friends are brilliant for a lot of work. Here's where a single conversation shines, where it starts to creak, and what changes when you give the work a team.
- →
Restless vs Paperclip.
Both are open source and both talk about running a company with AI agents. They start from different questions, and the difference matters more than the feature list.
- →
Restless vs OpenClaw.
OpenClaw is a personal AI assistant that lives in your chat apps and on your devices. Restless is a shared office for a team and its agents. Different shapes for different jobs.
- →
Restless vs Lindy.
Lindy is a hosted AI teammate wired into your email, meetings and team tools. Restless is a self-hosted office where your team and agents work on the output together.
- →
Restless vs n8n.
n8n is brilliant at explicit, repeatable automation. Restless is for open-ended work where judgement matters. They're more complementary than competing.
- →
Restless vs Dify.
Dify is a platform for building AI apps, RAG pipelines and agentic workflows. Restless is a place for a team and its agents to run work. One builds the tool, the other is where the work happens.
- →
Restless vs building your own with CrewAI, LangGraph or Microsoft Agent Framework.
Agent frameworks are toolkits for developers building their own agent apps. Restless is an app you use. Here's how to tell which one you actually need.
- →
Restless vs Manus.
Manus is a polished AI workspace for handing off tasks and building things with an agent that can use a computer. Restless is an office for a team and its agents. Here's how they differ.
- →
Restless vs Relevance AI.
Relevance AI builds teams of agents that run a business process end to end, wired into your existing apps. Restless is an office where people and agents work on the output together.
- →
Restless vs OpenHands.
OpenHands is a self-hosted control centre for coding agents. Restless is a self-hosted office for a whole company's work, code included. Close cousins, different jobs.
- →
Intelligence scales. Organisations still have to exist.
Better models won't melt companies down into one magic agent. If anything, they make responsibility, memory and authority matter more.
- →
Human-in-the-loop runs out of human.
Approving every step doesn't keep you in control at scale. It burns through the attention you need for the calls that actually matter.
- →
Failure is a capability.
Reliable agent systems don't prevent every failure. They make attempts throwaway while keeping the responsibility, the evidence and the way forward.
- →
Context management is management.
The hard part isn't cramming more history into a prompt. It's deciding what each worker should know, trust, ignore and carry forward.
- →
Stable structure beats intermittent brilliance.
A company can't rely on every model call being amazing. It needs identity, memory and responsibility that survive ordinary bad days.
- →
The agents could work. The context couldn't travel.
Multi-agent systems often fail at the handoff, not the task. Our own runs show how duplicated, missing and made-up context ruins perfectly good work.
- →
A critic that thinks like the worker is barely a critic.
Adding a reviewer doesn't magically create independent evidence. Useful criticism needs a different job, a different frame or a different way of looking at the result.
- →
One plan is not a strategy.
Useful autonomy means being free to branch, test competing ideas, drop the weak ones and bring in different skills as reality changes.
- →
The runtime is a subsystem, not your laptop.
Giving an agent shell access is easy. Giving a company a controllable, lasting, replaceable place to work is a design decision.
- →
Receipts scale where rules can't.
You can't make complex work safe by writing down every good decision in advance. A few hard boundaries plus solid receipts keep you flexible and accountable.