AI Agent Build Notes
Lessons from running agents on our own accounts.
Field notes and guides from building agents and running them in production. Where a note carries a figure, the figure has a footnote and a source.
All notes
Instagram DM Automation Without the Risk: Five Rules
Five rules from running an Instagram DM agent on our own account: approved answers only, money stopped in code, owner override, no duplicates.
Best AI Chatbot for Small Business: Choosing by Type
Four kinds of chatbot compared on what they do, who sets them up, what happens when they do not know, and who holds the data.
What Are AI Parameters, and Does the Count Matter?
A parameter is a learned weight or bias. What the count decides about memory and running cost, and why it says little about quality.
AI Model Quantization Explained: Smaller Models and What They Lose
Rounding the learned weights cuts memory and speeds replies. Where the loss shows first, and how to judge a compressed model.
Why AI Models Need GPUs: Cores, Memory and Bandwidth
Billions of multiplications per word and a full read of the model from memory: why bandwidth sets the speed, and why renting wins.
Not sure what you need? Ask in writing.
Describe the work in a few lines. We will reply in writing within one business day with what we would build and how.