japhyr

I've only implemented support for one provider at the moment, but it is structured in a way where I can add support for any other providers or local models. It feels good to not be locked into any one company's models or harness.

I'm most familiar with Python, so I built a Textual app using an OpenAI API plan.

There's just two boxes to start with: an input message box for prompting, and a conversation box. Dump each input message into the conversation box. Post the response to the conversation box.

Once you get a prompt-and-response cycle going, you can start to define tools. A tool is a description of an action the model can request to carry out, and you write a function to implement that tool call. The tool descriptions get sent with each API call, and then the model can start calling tools as part of its response.

I added a box to clearly show the project directory, and a couple boxes to show which tools were being called and whether the model was attempting to cross any boundaries it shouldn't. A fun milestone is when you get to use the harness you're building to help build itself.

I didn't want to over abstract the implementation, but it's reasonably well structured. The TUI is separate from the tool definitions. The tool definitions are model-agnostic. Everything specific to OpenAI is in an openai_model.py file. If I want to use Anthropic models, I could add an anthropic_model.py file. Or I could make it all local and add a local_model.py file.

I'm not working with subagents yet, but I can see a way to implement that if I end up wanting to.

The OpenAI Python library makes these interactions really straightforward to implement:

https://github.com/openai/openai-python

Yes, it's so easy to build a harness for a lot of use cases now, and you learn so much from doing so.

I'm a solo developer, and I didn't want to sort out all the boundaries and permissions of Claude or other heavy harnesses. All I wanted was something that can read, write, and modify files in a project directory. I gave myself three days to see if I could make a harness that was complete enough to work on my own projects. It was usable in less than a day.

I love using it, because I know exactly what its boundaries are, and how they're enforced. When I need more features I can either implement them in my own harness, or choose a heavier harness with much clearer understanding of what to look for.

When people have gotten excited about having so many tests, I've asked them what their plan is when 500 of their 1000 test suddenly stop passing. I've always favored 50 tests I know are meaningful, where any failure is something to respond to, over 500 tests with a bunch of overlap and edge cases I'm not sure matter. But now people aren't even thinking about that, they're just letting the agents make all those decisions.

I think this approach will continue to work for some people and orgs, but others are going to hit walls where their agents can't manage the mountains anymore. It's going to be interesting to see how people and orgs respond when that happens.

He doesn't really think, he reacts. And his every reaction is centered around "How can I make this about me?"

I like this article. We've read a lot of takes quite similar to this, and I think it's worth continuing to write and share reflections just like this.

> Yet, through much effort and 40 years of debate we had reached an equilibrium. Now AI has thrown that out the window.

The only thing I'd like to add here is that "AI" in this sentence should really be read as "the companies building frontier AI models have thrown this out the window". I don't necessarily think the author should have used a phrase like that, because it gets quite pedantic to read posts where such precise language is used all the time. But when we talk about where to go as an industry and as a society, we need to keep that distinction in mind.

The tech is here to stay, but the tech itself is not what's thrown our norms out the window. It's the people running these companies who are putting their vision and their profits above everyone else in the world. Most of us are exploring this tech and trying to figure out what it's useful for. It's a much harder question to try and figure out what to do about the people using this tech to consolidate wealth and power. We can't hold a technology responsible for anything, but we can hold people responsible for their impact on everyone else.

I'm the author of Python Crash Course, and I got this exact same email this week. I was thinking of writing a public response as well, because any attempt to sincerely answer these questions takes something along the lines of a full post. It's also worth a public response because many people who are getting into programming for the first time right now are asking variations of these same questions.

> Do I think that AI enables people to develop faster than they can keep up?

Absolutely. That's the core of this person's email, and everyone else who asks similar questions. Just five years ago, the only way to build a working project of moderate complexity was to learn the basic to intermediate concepts required to make an MVP. Now, if you can steer an LLM reasonably well, you can quickly build an MVP that goes well beyond your own understanding of the implementation.

I don't think anyone has clear answers to all the questions brought up in this email. I think people can learn faster than they used to, because they can make connections between different areas faster than they used to. But it requires skill and discipline in how you learn, and how you work. You have to intentionally build your understanding as you build your projects.

Are you seeing companies that are in over their heads because they don't have enough domain knowledge?

I think that's always been an issue in tech, but it feels like AI exacerbates the issue. I see people starting companies and services where their primary domain knowledge seems to be whatever their AI tells them. It's sometimes enough to get off the ground, but I wonder how long it takes for customers/clients who actually need the domain knowledge to bail.

For a bit of context, Juneau Alaska deals with a smaller scale version of this most years. There's a bowl off of the Mendenhall Glacier named Suicide Basin. Each summer that basin gets filled with meltwater, which forms a lake because the glacier blocks the water from leaving the basin. At some point the lake fills with enough water that part of the blockage gives way and the lake suddenly releases.

In Juneau, this is a well-defined basin with well-established dynamics. The city still wrestles with how best to handle this. Recent years have seen higher river levels when the basin releases, and the downstream banks of the Mendenhall River have eroded much faster than people were expecting. The city and individual homeowners are engaged in multiple projects to manage the annual flooding.

In the Nepal case the scale is much larger, and the variables are less known and less stable. This is absolutely devastating to see.

https://www.adn.com/alaska-news/2026/08/12/juneau-officials-...

> I did a full (human) blank page rewrite three days ago

You did a "blank page rewrite" of a 576 page book "three days ago"?

Of course everyone hates AI. For most people, it's actively making their lives worse, not better.

AI is a really effective technology when used with restraint, in appropriate places. But it's rarely being used with restraint, and it's being inserted just about everywhere.

If you're trying to work in AI on a product related to education, you've got a really high bar to clear to prove that you're actually helping students learn better, and teachers teach better.

Where are you located? Around 2020 I put together a landslide risk monitoring site that's entirely based on stream gauges:

https://khmpsitka.org

It's an interesting approach that's way simpler than relying on soil moisture sensors placed high up on mountainsides. It's not as widely applicable as measuring soil moisture directly, but it's been running steadily for over 5 years now.

I didn't see this until this morning, and the continuing through for the whole 18 words was a really good experience. I got 14/18, and that was much more satisfying than ending at my first failure.

Growing up in the 80s, I've known (and sung) this song all my life. But I just saw this flowchart for the first time today:

https://jeannr.tumblr.com/post/165291081/i-made-a-flow-chart...

I believe that's the original source, but it looks cut off. Here's a full version:

https://i.pinimg.com/originals/b1/82/dc/b182dcc291495c013c98...

When vehicles were starting to get noticeably bigger about 20 years ago, I asked some family members why they were buying vehicles they had to climb up into.

"On the highway, I'd rather kill someone than be killed in a wreck."

They would not recognize that while that might work for a while, it wasn't going to lead anywhere good for our society. A generation of people thinking like that has filled our roads with vehicles that protect their occupants while making it more dangerous for everyone else.

> In fact, I think Guido himself resigned due to the experience he had trying to get a PEP through the committee.

If you're referring to the steering council, that group was created in response to Guido stepping down.

He stepped down partly in response to the changing nature of online discussions around changes to the language. He just didn't want to be at the center of every polarizing discussion anymore. I think he also recognized that the transition needed to happen at some point, and that was as good a time as any.

> But adoption is the starting line, not the scoreboard.

Yes yes, shout it from the rooftops! Over the next few years I think we're going to see that companies that get this point will keep doing meaningful things, and stand a chance of weathering this transition period.

I think we're going to see a bunch of companies that went all in on AI for AI's sake go under because they've lost their mission, lost their implementation, and won't have a way to get those back in a reasonable timeframe and at a reasonable cost.

It depends on your definition of "threaten", but the short answer is yes. I live in a mountainous area that's a destination for road bikers and mountain bikers. There's lots of sources of tension between cyclists and drivers.

On a regular basis I see people racing past cyclists, rolling coal at cyclists (I can't believe that's even a term now), blaring horns, and a number of other behaviors that fall under "threatening".

US vehicles, especially pickups, have outgrown a lot of rural roads that had their origins as footpaths and horse paths. Even with well-intentioned cyclists and drivers, it's often times a setup for conflict.

So many areas in the US are much less walkable and bikeable than they used to be. I say that as someone who bicycle commuted for years. When I rode my bike to school as a kid I dealt with 25-35 mph traffic. The traffic was much lighter, the vehicles were much smaller, the drivers weren't perfect but they were way less distracted, and the shoulders were in better shape.

We can try to raise our kids with values that are consistent with the ones we grew up with. But trying to give them the same conditions because "it's what we did" doesn't always match up with reality.

This is great, but it's also easy to go too far in this direction. This can work through elementary school and into middle school, but I don't think it works in high school.

It's really hard to be a high school student without your own phone. I know some people who have kept their kids from having phones into high school. It avoids some of the addictive and distracting issues that come from having phones at a young age, but it's way more isolating than people realize. You might have a landline, but if no other high school age people are making voice calls to communicate, no one's going to call that landline. And the landline at home doesn't help you coordinate pickups and drop-offs as people start to do a wider variety of activities.

We have plenty of conflict in our home around devices, so I don't criticize any particular approaches. I'd just say that if you're taking this approach, it's probably a good idea to figure out how you're going to transition to kids having devices as they get into their high school years.

Have you ever met Chad? I have no doubt this is sincere.

I met Chad during the gittip years, and one of my life goals now is to go find him in the offline world some day and sit around an open fire at night to share stories. :)

I wish I still had my gittip penny, but I seem to have lost it in several moves since that time.

Anecdotal data point: My son is finishing 9th grade, and he's taking 10th grade math because he got ahead a year when he was younger. At his school, you're exempted from having to take the final exam if you're passing with a reasonable grade at the end of the semester. He said there are about four students who don't have to take the final exam.

Math has always been hard to teach well, because issues with earlier math classes compound so much. With all the societal interruptions to education, and the impact of addictive tech on young people's minds, it's only gotten more difficult.

I'm partway through Anthony Shaw's NDC talk, "Are LLMs good software engineers?" One of the realizations he shares is that he found himself treating AI assistants like junior engineers. Then he realized they're like junior engineers in how they work and behave, but they don't learn like juniors do.

I thought that was an interesting thing to point out.

People are talking about Covid, smartphones/screens, social media, and AI. No one has mentioned defunding public education yet.

In Alaska, where I lived most of the last 20 years, education has been largely flat funded for about a decade now. Imagine running an organization in 2026 on that organization's 2016 budget. Schools have a bunch of obligations they have to spend on. Every time health care costs for staff go up, and funding is flat, something gets cut. You can't cut education for a decade straight without impacting student learning.

I don't think Alaska is that much of an outlier in this regard.

> With most parks the crowds quickly disappear once you are 2 miles in.

And those crowds just aren't present in the early morning. Nobody gets up early enough to be out at sunrise. I'm not worried about saying this out loud and spoiling it either, because most people just don't like to get up that early.

I've been to gorgeous places all over the US that are absolutely packed by 10am or noon. Those same places are completely empty, and even more beautiful at sunrise. I live near one of the best mountain biking places in the southeast US, and regularly do 20-30 mile rides starting at sunrise, and only occasionally see a runner or another cyclist. There's just nobody out in the early mornings.

From my understanding there are a lot of companies that need their own package repositories, for a variety of reasons. I listened to a couple podcasts where Charlie Marsh outlined their plans for pyx, and why they felt their entry into that market would be profitable. My guess is that OpenAI just dangled way more money in their faces than what they were likely to get from pyx.

Having a private package index gives you a central place where all employees can install from, without having to screen what each person is installing. Also, if I remember right, there are some large AI and ML focused packages that benefit from an index that's tuned to your specific hardware and workflows.

This has me thinking about VS Code and VS Codium. I've used VS Code for a while now, but recently grew annoyed at the increasingly prevalent prompts to subscribe to various Microsoft AI tools. I know you can make them go away, but if you bounce between different systems, and particularly deal with installing VS Code on a regular basis, it becomes annoying.

I started using VS Codium, and it feels like using VS Code before the AI hype era. I wonder if we're going to see a commercial version of uv bloated with the things OpenAI wants us all to use, and a community version that's more like the uv we're using right now.