I agree, I tend to give a high-level concept for what I want done first, and then break it into smaller tasks. That said, there's almost always a misunderstanding within the smaller tasks anyway, which I have to test to find out about, then ask the agent to fix.
Thanks, yeah I just went with the most controversial term as the catch-all which just means: you prompt an agent as your core workflow.
For me as well, I do projects for clients that would have taken me months, and I deliver them in a fraction of the time. It's amazing and totally helped cure my burnout (which I had for over a year before I found Claude Code.)
I'm making the case that we should make tools for the higher abstraction, not a lower one. You said it yourself: why should we read/write binary? We don't, we shouldn't. It might be useful in some cases, but working with assembly is much easier.
Intent is more natural to us than code, for the same reason assembly is more natural than binary.
I never said you have to turn your brain off to feel the limitations.
Actually, for more complex work I think it's pretty common to spend a long time crafting some elaborate prompt, and then arguing with the agent for 10-20 turns, getting a "passable" plan, accepting it then arguing with the agent every step of the way because it's doing it wrong. It's incredibly frustrating, even with models like Fable.
Even after that, I'll often find out during later sessions that some feature that was supposed to be deprecated was actually silently left in "because I wasn't sure you wanted it completely gone" or something.
I agree that exploring a codebase through prompts is actually quite nice! But the mental model you get from that is almost always warped - you have to supplement that with reading the code, where 9 times out of 10 you will find some kind of discrepancy that you're unhappy with.
The control room in Jurassic Park was honestly beautiful - and something about those shaving cream cans really crystallized the aesthetic for me, weirdly. I had no idea that it was taken so seriously, I just remember thinking "damn, that's actually pretty realistic" when I re-watched the movie a few years ago.
It's still pretty experimental, but I think now that AI can feasibly generate code from intent/specs it's worth doing... especially with how much vibe coding is both a blessing and a curse.
Basically, what if we took those markdown specs everyone likes using and systematized them into a model graph, anchored claims to code, then represented that as a wiki? And what if we surfaced a planned change to a codebase as a diff of the definition of that model? You can then properly keep track of what gets implemented and how, because you're no longer reading an essay from the agent as a plan, you see a complete diff of everything it plans to do.
I've been using it myself for work, but it'd be nice to get some feedback from other people's perspectives.
This is all from the perspective of people who already oppose Farage, though. Supporters will zealously reframe everything as illegitimate attack. Soft skeptics will notice the establishment's overreach.
I agree this might be a normal political tactic on paper, but that's irrelevant when you're dealing with populism. The entire game is tribal framing. To anyone outside the anti-Farage coalition, this just looks like the establishment closing ranks, and that's more recruiting material for him.
What I can see happening here, at best, is maybe Farage stepping away if the pressure gets too much. But then you risk him getting replaced by a more angry figurehead who will use Farage's downfall as evidence of the "hatred and dirty politics" of the establishment.
I lived in London for six years starting just before Brexit, so I watched this exact confidence up close. The British public "knowing" Farage is damaged is not new, it's not justified, and it doesn't do anything except make people feel better on social media.
I'm pretty sure that just like every other backhanded attempt to squash the populists, this will backfire. I can't know that of course, but I'm pattern-matching against years of this kind of stuff happening in like 5-6 different countries in Europe already.
This doesn't make a fool of Farage though! It clarifies to anyone already skeptical of the establishment that when pushed, they'd rather not play than risk losing. This is literally self-defeating!
You claim that this kind of thing "generally harms populists more than anything else," but I can't think of a single time that it actually has at the scale that matters. What would harm populists would be a viable alternative that takes the concerns of the population seriously.
I don't understand why the mainstream is celebrating the total collapse of any meaningful political opposition to populists. This kind of fluffed-up 2010's style optimism campaign is exactly how the establishment lost control in 2016 in both the US and the UK.
I think it's sad that Binface is the only one even attempting to engage in democracy at this point. He's a satire but he's also, seemingly, the only legitimate opposition that populists face. That should be concerning and embarrassing, but no one has learned anything over the past 10 years.
So, what happens when Binface loses? Are they just going to try to put Farage in prison like they're doing with the other populists? That'll surely help.
And if he does somehow win, what then? Doesn't that also prove that populism is the only viable strategy? Either way people are celebrating their own defeat.
RE: Tauri not having cross-compile... There's a GitHub action that compiles for Linux, Windows, and Mac. So practically it does have it, just not out of the box.
Just started watching Widow's Bay, which to me has a similar vibe, but a much better hook: folk horror as the plot driver.
I can't watch Northern Exposure anymore. It's too alienated from the modern world, it feels like an artifact from a different civilization rather than anything I can relate to now, which is sad.
They're in every discussion even remotely related to anything Elon Musk.
Everything Elon does is somehow stupid or evil. Actually, that reminds me of Thunderf00t YouTube streams where he was (or still is?) betting Starship would fail miserably every test flight, and he'd talk about how evil and stupid Elon is for 3 hours with chatters, watch the flight then say something like "it's still bullshit."
I think it's a mixture of cope and a little bit psyop from adversaries like Russia who are being crippled in Ukraine because of Starlink.
I felt this way from a year ago up until February 2026. Claude Code and Codex becoming the norm cemented for me that a lot of the projects people are working on (including mine) are totally obsolete. As far as I'm concerned, most code is now abstracted away, and people only want better agents - not traditional software products, except as infrastructure or platforms.
It also looks like the final form of the AI roll-out: whatever the model or application, this is the era of agents, and probably in the near-future mostly automated agents. We'll see an overflow of bespoke automation and in-house agents doing everything from personal task management to enterprise business processes, so releasing a "Personal Fitness Tracker" or a "CRO Auditor" in 2026 doesn't make any sense.
All of my anxiety around it has evaporated because I can see what it actually is: an ouroboros of AI output generating automation of more AI output. What most software engineers will be working on now is guiding that output, making it easier to inspect/configure it, optimizing it, and improving the consumer and developer experience.
Otherwise, we just have to drop our old concepts for projects and work on something else.
For the consumer the floor is rising, and for the experienced developer the ceiling is rising. I personally hate web dev anyway, and I'm glad I can work on interesting engineering problems (even with the help of an AI) instead of having to manually stitch together yet another REST API, or website, or service pipeline.
I don't get it, you joined this thread to call me an idiot with a meme, and now you're talking about being a neutral arbiter for a technical discussion that I supposedly ruined.
More than anything I'm getting frustrated with HN discussions because people just insinuate that I'm stupid instead of making substantive arguments reasoning how what I'm saying is wrong.
Are we performing for an audience or having a discussion?
Just give us a taste of what we'd be paying for? I'm sure you're an expert but before I commit to 2+ years of consultation I'd like to see your approach.
This is a design problem, and not something necessarily solved by CLI --help commands.
You can implement progressive disclosure in MCP as well by implementing those same help commands as tools. The MCP should not be providing thousands of tools, but the minimum set of tools to help the AI use the service. If your service is small, you can probably distill the entire API into MCP tools. If you're AWS then you provide tools that then document the API progressively.
Technically, you could have an AWS MCP provide one tool that guides the AI on how to use specific AWS services through search/keywords and some kind of cursor logic.
The entire point of MCP is inherent knowledge of a tool for agentic use.