korrents
Kent C. Dodds

What Kent C. Dodds thinks about AI agents

@kent-c-dodds · 71 positions · On ppll ↗

Software engineer and educator; creator of Testing Library and of the EpicWeb.dev and EpicReact.dev courses.

Kent C. Dodds did not write this page.

We collected these quotes from things they published elsewhere, and every quote links to where it was said. They have no account here and have not endorsed this site. Quotes are word for word; the short line under each one is our own restatement, not their wording. Their own site. Is this you? Claim it or ask us to remove it. Or tell us what is wrong here.

27 dated positions, 2026, in their own words. Our reading of what Kent C. Dodds has said — not written or endorsed by them.

  1. The secret to the software factory is getting agents to complain about what they hate about working in your product/codebase.

    ↗@kentcdodds on Xx.com

  2. 9 days earlier
  3. ADRs and primitives are a huge leverage point for getting your agents to build the right thing.

    ↗@kentcdodds on Xx.com

  4. 13 days earlier
  5. Unbelievable amount of leverage with agents doing the implementation.

    ↗@kentcdodds on Xx.com

  6. 4 days earlier
  7. Agents are going to eat the entire software development lifecycle

    ↗@kentcdodds on Xx.com

  8. 5 days earlier
  9. But they won't ask for an app, they'll just ask for a problem to be solved and the agent will write the code under the hood to solve it.

    ↗@kentcdodds on Xx.com

  10. I've not used it in a long time because agents are good at managing that on their own

    ↗@kentcdodds on Xx.com

  11. 2 days earlier
  12. preventing it, is a pretty simple 3 pronged process: 1. Ensure agents create high quality PRs in the first place 2. Have good agent PR reviewers to catch problems before they get to prod 3. Use feature flags and have a (regularly tested) disaster recovery process if a problem manages to make it to prod

    ↗@kentcdodds on Xx.com

  13. 6 days earlier
  14. You should look for ways to turn what your agent does into deterministic software.

    ↗@kentcdodds on Xx.com

  15. 1 day earlier
  16. Like, keep your mind and your eye on what's going on uh outside of what you're typically doing because you might know something or or think of something they hadn't thought about before that you could share with your agent and your agent can look into it.

    ↗Steal My App (I'll Show You How)youtube.com 1st of 2 in this recording

  17. 7 days earlier
  18. Most modern agents support browsers and that's fine most of the time.

    ↗@kentcdodds on Xx.com

  19. Each one of these products (and the others inevitably on their way) are going to do their best to make theirs the winner. But the fact is that they will continue to leap frog each other in capabilities.

    ↗@kentcdodds on Xx.com

  20. But the thing is, they all want to drive their claws so deep into the flesh of your business (to be useful of course), that it will be a huge chore to get rid of them. The integrations, automations, memories, secret tokens, etc etc etc. All of these will be locked into the assistant and they are not incentivized to make it easy for you to transfer these things to your "new hire."

    ↗@kentcdodds on Xx.com

  21. 5 days earlier
  22. Agents are really good at output now, and they're getting better fast. What they're not good at is knowing which output is worth producing in the first place.

    ↗You Own the Outcome: Product Engineering in the Age of AIkentcdodds.com 1st of 2 in this piece

  23. 8 days earlier
  24. I don't yet have numbers to back this up, but I gotta believe that someone connecting their agent to Kody in one agent and then switching their main driver to another somewhere else is suboptimal.

    ↗@kentcdodds on Xx.com

  25. 1 day earlier
  26. Uh, actually having AI write a first draft of an email is pretty nice, but having it send that email without you reviewing it can be irritating.

    ↗Make Your Agent Safe and Autonomousyoutube.com 1st of 9 in this recording

  27. If you watched my video the while back, you saw how an agent was able to drop a production database because it was an overscoped token.

    ↗Make Your Agent Safe and Autonomousyoutube.com 2nd of 9 in this recording

  28. And this is a a general practice that I would recommend is that you can just ask the agent if you ever get stuck on anything. Just tell the agent that you're stuck and ask it what to do next.

    ↗Make Your Agent Safe and Autonomousyoutube.com 3rd of 9 in this recording

  29. and the more and more capable you make your agent, the more it can do kind of on its own and you don't have to control every piece of what it's doing and and be the like context hot potato person.

    ↗Make Your Agent Safe and Autonomousyoutube.com 4th of 9 in this recording

  30. When you run into a problem with your agent, you can literally just take a screenshot here. We'll we'll do the full thing right here. Paste it into your agent without any context at all. Just here's the screenshot and it will figure out what's going on and tell you what you should do next.

    ↗Make Your Agent Safe and Autonomousyoutube.com 5th of 9 in this recording

  31. Yes, we trust the agents to act in good faith and stuff, but sometimes they misunderstand us. And if they have the power to drop your production database, that has happened and it could happen.

    ↗Make Your Agent Safe and Autonomousyoutube.com 6th of 9 in this recording

  32. So you want to make a package of deterministic code that can only do the things that you know you want them to be able to do. And then you can lock down the integration and lock down the package so that agents aren't able to do more than the things that they should be allowed to do.

    ↗Make Your Agent Safe and Autonomousyoutube.com 7th of 9 in this recording

  33. I want you to take a look at your&env file and look at all those tokens that your agent is happily slurping up and able to use and uh figure out how to get those integrations into a safer place where agents won't be able to just willy-nilly use tokens uh however they want to

    ↗Make Your Agent Safe and Autonomousyoutube.com 8th of 9 in this recording

  34. And uh I just want you to be a little bit more intentional about what your agents are able to do so that you can reduce the chances for catastrophe.

    ↗Make Your Agent Safe and Autonomousyoutube.com 9th of 9 in this recording

  35. You ➡️ agent ➡️ software This is the path to efficient processes and too many of you stop on the first two steps costing money and time.

    ↗@kentcdodds on Xx.com

  36. 5 days earlier
  37. Keep finding ways to climb that abstraction ladder! Your agent wants your job! Give it!

    ↗@kentcdodds on Xx.com

  38. 6 months earlier
  39. coding agentsdesign

    AI agents make redesigns cheap. The reason I could go through three rounds of "build, ship, regret, redesign" in a matter of days is that the agent did most of the implementation work.

    ↗Implementing Hybrid Semantic + Lexical Searchkentcdodds.com 2nd of 2 in this piece

  40. 5 days earlier
  41. structural refactors break assumptions you didn't know you had, and "the agent was confident it would work" is not the same as "it will work."

    ↗Simplifying Containers with Cloudflare Sandboxeskentcdodds.com 2nd of 3 in this piece