One of my favorite AI How-tos in the last year. Barry and Erik spend 80% of the post saying ~”eh, you probably don’t need agents. Just build straightforward deterministic workflows with if-statements instead.”

And then, when you actually do need agents, don’t over complicate it!

This post also introduced the concept of an Augmented LLM — a LLM hooked up to tools, memory, data — which is a useful abstraction for evolving LLM use beyond fancy autocomplete.

“An augmented LLM running in a loop” is the best definition of an agent I’ve heard so far.

I have ADHD, and my daughter was recently diagnosed as well.

Been reflecting on this post as it's been soaking up the front page for the last 24 hours.

I want to commend you for shipping maybe the perfect HN post:

- Personal Journey

- Old school hardware

- DIY software

- Productivity hack

- Great title

- Quantified results (2-3x productivity) over non-trivial duration (few months)

- A low-effort offline solution that delivers real value for the 98% who will never build the thing

- Great polish on the reading experience with lots of little details

- Effective call to action (subscribe to get the software in a few weeks)

You inspired me to get my organization back on track. After researching receipt printers for 30 minutes, I realized what I actually need is to dust off the system that has worked for me in the past. But I'm picking up some post-its today and my daughter and I are going to try implementing your system for her over the weekend.

Thank you for putting the time into this!

Unlimited Claude Code for $100 on the Max plan is a game changer.