I am still stunned that there are people who hate AI so much that they have a problem with the weights of an LLM being on their computer. To me, that sounds rather esoteric.

Batteries help, but even that is limited in northern countries like the UK. If you look at the data, in July '25, solar produced 2.36 TWh. But in December '25, it was only 0.535 TWh: the output in summer is >4 times the winter output. So either you need to discard 75% of the electricity produced in summer, or you need truly gigantic batteries that store power produced in summer for winter. Both is not economical. Solar is far less efficient in the UK than in, for example, Florida.

The reason why personal solar installations are profitable is that you can buy electricity for fixed prices from your local power company. You pay the average of the vastly different low (or negative) prices during the day and the extremely expensive prices on windstill nights. Solar allows you to use your own electricity when the average is below spot prices, and get power for much less when the price you pay is cheaper than spot prices. It's like a state-approved scheme to play the market in the name of decarbonization while actually increasing everybody else's prices and possibly even CO2 emissions.

I don't know where the ceiling is. And how much of the improvement was due to better context engineering, and how much to better models. I would expect the context engineering to plateau very soon. Not sure about the models.

An even more dramatic change for the whole economy will be when non-IT, non-creative office clerks are replaced. This is mostly a matter of redesigning the interfaces around them. AI could probably do already most of the work, but getting the tasks to the AI, using their output, and communication with third parties are still a major challenge. Like someone processing insurance claims. AI needs a way to get the claim, to contact third parties (write emails to humans, communicate with other AI agents, maybe even call humans), and then to initiate the payout. It's already doable with today's technology, but still a lot of work.

True, but junior developers used to provide a lot of value while doing this. Now their value, while they are still figuring it out, has gone down immensely. For a company, there is no value in letting a junior dev write code anymore. And for reviewing the AI output, you need someone more experienced.

Simplicity. I can get you reading your first feed in under a minute. Also, I am not really thinking about monetization right now, but I am building a feed reader I want to use. I wouldn't want to spend $13 a month for it.

> thin UI + OSS pipeline

No, the UI isn't that thin. I am optimizing it to minimize my costs for operating it. Everything I can do inside the client is done inside the client. Interactions with the server are mostly limited to polling every 2 minutes for feed updates, and sending read markers after 3 seconds of inactivity. Feed data is stored on CDN, compressed.

Yes, enterprise is certainly where the money is (Feedly's plans start at $1600/month...), but as a solo dev working on a side-project, that's not an accessible market for me anyway. So I try to create a service that's simple and cheap.

> My AI agent is (probably) going to do a better job of choosing the most relevant stuff.

The idea would be basically: the feed reader know the user's interests because of the subscriptions, and knows the last time the user logged in. So it can filter what happened since then; it can also order the posts by relevance, allowing the user to catch up. And in a second step, an agent could even write the posts dynamically, summarizing information gathered from the user's feed, possibly even adjusted to the user's level of knowledge and offering background info where needed.

> And personally, less interested in podcasts in my RSS app. That's something for Pocket Casts / AntennaPod. I like my audio separate from my RSS.

There are some feeds that are more like a mixture of text and podcast. I usually read only the text, but sometimes it catches my interest and I want to listen to one or two posts. That's when I start hating the lack of podcast support in Feedly.

For people who do not want to use self-hosted services (which generally includes me), it offers simplicity. Open the page, choose Google as auth provider, confirm, and you will get a friendly start page. Click on 'follow' on one of the feeds, and you can start reading immediately. The UI is more like Facebook or X, so basically, you just need to scroll. Either in a feed of your choice, or all your feeds. It's designed to work well on small mobile screens, tablets, and desktops, with great keyboard support on the latter. Larger screens use two or three columns.

I have plenty of other ideas for what to build on top of it: offering an SDK and APIs so you can vibe-code the UI you want, a built-in podcast listener, using news from aggregated feeds to build a personalized AI feed. But the first step is to reach the Google Reader feature set minus social features.

Admittedly, I have never tried to run it that long. If 10 minutes are not enough, I check what it is doing and tell it to do what it needs to do differently, or what to look at, or offer to run it with debug logs. Recently, I have also had a case where Opus was working on an issue forever, fixing one issue and thereby introducing another, fix that, only for the original issue to disappear. Then I tried out Codex, and it fixed it at first sight. So changing models can certainly help.

But do you really get a good solution after running it for hours? To me, that sounds like it doesn't understand the issue completely.

I don't think there will be a future where agents need to work on a limited piece of code for hours. Either they are smart enough to do it in a limited amount of time, or someone smarter needs to get involved.

> This can't be a serious project. It must be a greenfield startup that's just starting.

I rarely review UI code. Doesn't mean that I don't need to step in from time to time, but generally, I don't care enough about the UI code to review it line-by-line.

If I read stuff like that, I wonder what the F they are doing. Agents work overnight? On what? Stuck in some loop, trying to figure out how to solve a bug by trial and error because the agent isn't capable of finding the right solution? Nothing good will come out of that. When the agent clearly isn't capable of solving an issue in a reasonable amount of time, it needs help. Quite often, a hint is enough. That, of course, requires the developer to still understand what the agent is doing. Otherwise, most likely, it will sooner or later do something stupid to "solve" the issue. And later, you need to clean up that mess.

If your prompt is good and the agent is capable of implementing it correctly, it will be done in 10 minutes or less. If not, you still need to step in.