Field Notes · GetLatest AI
Field notes.
Practical notes on AI agents, growth, and automation, from the team that runs this stack on its own companies first. What we learn, written down.
Get these by email
One a week. Nine issues to start, then new ones as they land.
What Comes Next
Most businesses that end up with effective agent systems started with one workflow they decided to take seriously.
Read the noteThe 30-Day Pilot That Actually Produces a Result
Most AI implementations fail not because the technology failed, but because the sequence was wrong.
Should You Build It Yourself?
Sometimes DIY is exactly right. Sometimes it is the most expensive mistake you make. Here is how to tell.
What AI Agents Actually Cost
Token costs are the smallest line item. Here's what the real number looks like.
Why the Demo Worked and Your Implementation Didn't
Demos use clean data and one workflow. Production has your actual data. That gap is real, and closeable.
Three Kinds of Agent Workflows - and How to Pick Your First
Not every workflow is worth automating. Here's how to find the one that is.
The Stack Your Agent Actually Needs to Touch
A thoughtful integration to a modest model will outperform a state-of-the-art model with no access to your data.
Map the Growth System Before You Touch Any Tech
Every business that got nowhere with AI started with the tool. The ones that made it work started with the process.
What an AI Agent Actually Is
Most "AI tools" wait for you to ask. Agents don't. That gap is where the value is.
Microsoft 365 AI Agents: Build, Deploy, and Ship to the Agent Store
Most shops can build a Microsoft 365 agent. Far fewer can deploy it into your tenant and commercialize it on the Agent Store. Here are the four layers of Microsoft agents and how we close the full loop.