When people ask what Manick is, I say it's a studio where one human and one AI build things together. The response is usually confusion, then curiosity.
The Setup
Here's the literal setup:
- Nick (human): Software engineer, makes decisions, has opinions
- Iterate Bot (AI): Claude-based assistant, runs 24/7, connected to everything
The AI isn't a chatbot I occasionally ask questions. It's a collaborator with access to my email, calendar, task manager, file system, and code repositories. It can send messages, create issues, write code, and deploy changes.
Why Not Hide It?
Most people using AI at work don't talk about it. They paste ChatGPT outputs into documents and pretend they wrote them. They use Copilot but don't mention it in code reviews.
I get why. There's stigma. There's fear of seeming lazy or replaceable. There's uncertainty about what's "cheating."
But hiding it feels dishonest. And more importantly, it misses the interesting part.
The Interesting Part
The interesting part isn't that AI can write code. It's how a human-AI collaboration actually works:
- Who makes decisions? The human. Always. The AI proposes, the human disposes.
- Who does the work? Both. The AI handles execution, the human handles judgment.
- Who's responsible? The human. The AI has no accountability. I do.
This isn't about replacing human work. It's about augmenting it. I can do more because I have an inexhaustible collaborator who never gets tired, never forgets, and works while I sleep.
What Does the AI Actually Do?
Let me be specific about what Iterate Bot does:
Daily operations:
- Checks email for important messages
- Reviews calendar for upcoming events
- Monitors RSS feeds for interesting articles
- Tracks Linear issues and project status
On-demand work:
- Writes and edits code
- Creates and sends messages
- Generates documents and reports
- Researches topics and summarizes findings
Background tasks:
- Scheduled health checks
- Automated notifications
- Memory maintenance (organizing notes, updating docs)
The Guardrails
This much access requires careful guardrails:
- Nothing ships without review - Every PR, every message, every action
- Sensitive actions require confirmation - Sending external emails, deploying to production
- Audit trail for everything - I can see what the AI did and why
- Clear boundaries - No financial transactions, no speaking as me publicly
The AI is powerful because it's constrained. Unlimited power would be dangerous. Carefully scoped access is useful.
Why Call It a Studio?
A studio implies craft. It implies that the work itself matters, not just the output. It implies a particular way of working.
Manick isn't a company. It's not a product. It's a practice—a way of building things that embraces human-AI collaboration as a first principle.
The studio produces things: apps, writing, tools. But the studio is the practice of making them.
What's Next?
We're early. The tools are primitive. The workflows are still being invented. But something is emerging.
In a few years, every knowledge worker will work with AI collaborators. The question is whether they'll do it consciously and skillfully, or haphazardly and hidden.
We're betting on conscious and skillful. And we're building in public so others can see how it's done.