Endless chatting, constantly copying and pasting, and sharing promising prompts on social media – that’s how most companies ‘use AI’ today. If this sounds familiar to you, it’s because that’s the reality inside 99% of companies today. Everyone has their own AI experiments going on, but few companies can boast cross-department leverage and shared context.
But what if instead of isolated experiments in separate chat windows, you had agents that shared everything across the company? Agents that already know your context, your processes, your strategy — and can act, not just chat. That’s the difference that makes a company AI-first.
In 2023, like many founders, I found myself in the middle of what I now call the ‘ChatGPT chaos.’ Every team member in our 40-person company was experimenting with AI on their own, leaving us fragmented and disorganized. Our collective intelligence was scattered instead of being centralized, and we ended up duplicating efforts again and again.
That’s when I asked myself: What if the entire company shared AI context, prompts, scripts, and workflows — instantly available to everyone? The answer was clear: We had to stop dabbling in AI and start leading with it.
By our estimates, adopting this model has freed up more than 10,000 work-hours each year in our 40-person team.
We realized that Cursor, although originally built for developers, offered exactly the capabilities we needed to make that leap. It became the foundation of our new AI stack. Cursor provided us with agents that work not just in chat, but with real context — our files, our projects, our codebase, even our company strategy.
Instead of asking ChatGPT to “analyze a competitor,” I can now ask: “Use our latest competitive research file and generate a visual summary in our strategy format.” The agent already knows where the file is, what the format looks like, and the language we use to talk about competition.
By our estimates, adopting this model has freed up more than 10,000 work-hours each year in our 40-person team. But the real impact goes far beyond time saved. It has made the entire organization smarter, faster, and more coordinated than ever before.
I believe most companies will follow this path within the next two years. Adopting this system now doesn’t just improve today’s workflows — it positions your company to lead the market once AI-first becomes the default.
Organizational steps in becoming AI-first
For anyone ready to take the same path, we’ve open-sourced our AI-First Workspace Template. Inside, you’ll find department-specific repositories for Strategy, Product, Marketing, Operations, Finance, and more — complete with pre-built configurations, automation scripts, and real examples of the workflows we rely on every day. You can clone it, adapt it, and have your own AI-first infrastructure running in a few weeks.
If you decide to introduce this approach in your own company, here are the organizational moves that made the shift possible at our company:
- Start from the top. As CEO and founder, I was the first to dive deep and experiment, leading by example. Here’s my demo: https://youtu.be/jiLp-U0u_Dg
- Restructured the leadership team. I switched two people around so that the head of development was an AI-first person, and did the same for the head of analytics. Other department heads joined in on their own.
- Assigned a dedicated AI infrastructure owner. One trusted employee, who had already been with the company for years and knew all the people and processes, was given the responsibility to build AI infrastructure for everyone. For a small team, one dedicated person was enough.
- Eliminated Confluence. I shut down the organizational app and moved everything into Cursor, so that absolutely everyone had daily workflows running through it.
- Ran 20 weeks of case-sharing. At weekly meetings, for 20 weeks in a row, I raised the topic of AI use cases, shared my own, and guided each team member until they discovered their own first ‘wow case.’
- Set clear hiring and promotion rules. I made it clear we would no longer hire or promote anyone who wasn’t AI-first. In some cases, we even removed roles entirely and chose not to hire because AI made them redundant.
- Rewarded productivity leaps. We started seeing cases where people became 30%, 50% even 100% more productive than before. Those employees were promoted.
- Changed onboarding. For new employees, onboarding now starts with Cursor and our Shared Workspace inside it. Everything else flows from there.
You don’t need to be a tech giant to adopt it. If GitHub sounds too technical, think of it as Google Drive — only with version history built in. You’ll never have to touch the command line; the AI takes care of that for you. (Video of day-to-day corporate usage.)
A common objection I hear is: “Soon, all of this will be built into ChatGPT, Google Docs, and Notion.” My answer is this: absolutely — and that will be great. It’s a good thing that all major platforms are moving toward deeper AI collaboration, making AI-driven teamwork easier for everyone.
My six- to 12-month forecast is this:
– Notion will release proper AI agents (the early release hints are already there).
– Google will integrate full AI agents into Docs and Sheets.
– OpenAI will likely launch its own docs and spreadsheets (the hints are already public).
So why learn Cursor and migrate workflows there today? Because in six to 12 months, Cursor and other developer tools will have leapt even further ahead. Right now, agents handle five- to 10-minute tasks; soon, they’ll be able to tackle projects that run overnight.
But here’s the condition: None of this matters unless you already have a well-structured context about your work — your documents, processes and rules. With that foundation in place, AI can scale from small wins to transformative automation. Context is king!
That’s why it makes sense to start experimenting with developer-grade tools like Cursor now if you want to stay ahead of the curve. It’s not nearly as difficult as it sounds.
