jbethune

The SaaS version has a short video demonstrating the UX: https://ninchi.ai

He's speaking facts. I have had the same concern about excessive cognitive offloading. As others have noted in the comments, AI tools when used correctly, can actually make us smarter and help us learn faster. But that requires a very particular usage pattern that is different from what I see with all the vibe coding going on these days.

I created a project called Ninchi to force myself to read my code and understand it. Recently I began also sharing it to see if there may be a larger need/opportunity. It's a small effort. We need to make a variety of efforts I think to encourage responsible AI usage before we end up drowning in slop.

Sick. Been meaning to replay Link to the Past.

This is extremely true. The instinct to fill every bit of downtime with a quick bit of doom-scrolling is very hard to kick. It's something I have made a point of working on; giving my mind space to just do nothing and let all sort of mental detritus process itself.

Been living in Japan for a while and have come to appreciate bonsai. There is a lot that goes into it. I love the concept of such an intricate thing being carefully maintained across decades and generations of people.

This was a bit word-salad-y but I share the same basic concern. I think more I worry about the tendency toward greater and greater cognitive off-loading to LLMs. My sister told me a story the other day about how she caught her plumber using chatgpt on his phone to fix an issue with her bathroom. I just think it's good for humans to know how to do stuff.

Great work. I like the approach of creating a schema to work with the OpenAPI spec.

I would second this. Would be very helpful. Had a fat finger issue a while ago that this would have saved.

Local urbex and exploration of 'haikyo' areas. Easy for me since I am in Tokyo and it's super walkable. I have taken to just getting on the train and getting off at random stations and walking in a random direction for a couple of kilometers. Every now and then I run into interesting abandoned buildings or neat shrines. Also makes for good exercise.

Saved. Very useful. Normally I just dig around the Github UI to see what I can glean from contributor graphs and issues but these git commands are a pretty elegant solution as well.

Anecdotally I can say this is true both in education and software development. The diversity of approaches and writing styles among junior developers used to be fun to observe and mentor. Now with everyone using AI coding agents there is a same-y-ness to people's work that makes it harder to see what the writer actually knows or doesn't know. A friend of mine who teaches high school English has said the same thing about student work.

Congratulations on the launch! Will definitely test this out.

I think this is a model issue. I have heard similar complaints from team members about Opus. I'm using other models via Cursor and not having problems.

Forked. Very cool. I appreciate the simplicity and documentation.

This is super cool. Wish I had this back in my sysadmin days.

I think there is a meaningful distinction here. It's true that writing code has never been the sole work of a software engineer. However there is a qualitative difference between an engineer producing the code themselves and an engineer managing code generated by an LLM. When he writes there is "so much stuff" for humans to do outside of writing code I generally agree and would sum it up with one word: Accountability. Humans have to be accountable for that code in a lot of ways because ultimately accountability is something AI agents generally lack.