Why trust this
Proof
You can judge how I think before we talk.
These articles come from building and running AI-agent products of my own. They are the problems a team meets once agents write code and talk to customers: how to trust what they produce, how to limit what they can touch, and how to run them day to day.
Everything here is free to read. It is writing and not a list of logos, so you can check it for yourself.
Verify
How to know that what an agent produced is right: tests, guardrails and review.
- Which guardrails does an AI agent need before it goes to production?A checklist of input, tool, output and operations guardrails to clear before real customers talk to your agent, with a pointer to the deeper guide for each
- How to build guardrails for AI AgentsPut checks before the model, around every tool and after every reply, so your agent stays useful when a customer, a bug or a prompt injection pushes it off course
- How do I test that my AI agent guardrails actually work?Keep a small set of attack and edge-case prompts, run them on every change and watch real traffic, so a guardrail is proven and not just assumed
- What does a custom guardrail for an AI agent look like in code?Write one input check, one tool check and one output check as plain functions, then see where a framework helps and where it only adds weight
- How do I set up guardrails for an AI coding agent?Limit what Claude Code, Codex or Cursor can read, run and push with permission rules, a sandbox, your own hooks and a review gate, so a wrong guess costs minutes and not your repository
- How do I fix flaky Rails system tests when using Cuprite?Most flaky browser tests come from timing, animations and shared state, and each one has a small fix that you can apply in Cuprite and Capybara today
- How to test in Chrome using CupriteCuprite drives a real Chrome over the DevTools Protocol, so your Rails system tests are faster, steadier and closer to what users see.
Control
How to limit what an agent can touch, spend and break.
Operate
How to build agents, run them day to day and get them in front of users.
- How to build your own AI AgentsBuild a pantry agent that reads receipt and meal photos from a shared folder and leaves a shopping list on your phone every Friday morning
- How do I structure marketing for an AI product?Split it into three channels, the product itself, email and search, give each one an owner and a number, and judge every channel by purchases instead of clicks.
Send a short note
Read something that matches a problem on your team? Send me a few lines about it.
Email me
or write to ashish.gaikwad.pro@gmail.com