It's not perfect and it's not quite finished yet. At the moment Canny.io has more features and is better known, but I prefer the simplicity of the tool and the price (10$ once and own it forever). It only has basic features right now, but feel free to request features you think it needs!
No, I'm paid much more to do much more than what I did in this simple task. Claude didn't even test the changes (in this case, it does not have the hardware required to do that), or decide that the feature needed to be implemented in the first place. But my comparison wasn't "how do I compare to Claude Code", it was "how does Aider compare to Claude Code". My boss does not use Aider or Claude Code, and would not be happy with the results of replacing me with it (yet).
I said that the AI literally does not have the hardware required to do the testing necessary. But ignoring that, automated testing is not sufficient for shipping software. Imagine shipping a website that has full test coverage but never once opening the browser. This isn't a fundamentally impossible problem for AI, but no amount of "good prompting" is going to get you there today.
I think I pretty directly addressed that point. Yes, it would be more expensive to hire me to do what Claude Code / Aider did, but nobody would be satisfied with my work if I stopped where Claude Code / Aider did.
They aren't necessarily saying it can replace you. They're saying that even though it's expensive, it's cheaper than your time (which can be better spent on other tasks, as you point out.)
The first half is correct, but the conclusions shouldn’t be ‘we’re replicating our software engineers with Claude today’, they’re ‘our experienced engineers just 10x their productivity, we’ll never need to hire an intern’
Productivity gains decrease exponentially after a few weeks as your engineering skills become rusty very fast (yes, they do, in 100% cases)
Thats the biggest part everyone misses. It’s all sunshine and rainbows until in a month you realize you start asking llm to think for you and at that point the code becomes shit and degrades fast.
Like with everything else “use it or lose it”
If you don’t code yourself- you will lose the ability to properly so it very fast, and you won’t realize it until too late
If you're using the LLM poorly. Many team leads spend very little time programming, and spend a lot of time reviewing code, which is basically what working with LLMs is. If the LLM writes code that you couldn't have written yourself, you aren't qualified to approve it.
I'm pondering where this "AI-automated programming" trend is heading.
For example: thirty years ago, FX trading was executed by a bunch of human traders. Then, computers arrived on the scene, which made all of them practically obsolete. Nowadays FX trading is executed by a collection of automated algorithms, being monitored by few quants.
My question is: is the software development in 2025 basically like what the foreign exchange was in the 2000s?
With industrialisation blacksmiths were replaced by assembly lines.
I'm sure that blacksmiths are more flexible and capable in almost any important dimension, but the economics of factories just made more sense.
I expect that when the dust settles (assuming that the dust settles), that most software will be an industrial product. The humans involved in its creation will be engineers and not craftsmen. Today we have machinists and industrial engineers - not blacksmiths.
Quality and quality assurance processes will become more important, I also expect optimised production processes.
I think a lot of the software ecosystem is a baroque set of over-engineered (or over crafted) steps and processes and this will probably be refactored.
I expect code quality metrics to be super refined.
Craftsmen don't usually produce artifacts to the tolerances that our machines do now - code will be the same.
I expect automated correctness proofs, specification languages, enhanced type systems to have a renaissance.
I know this is not really in the spirit of the room here, but before I ever dreamed of getting paid to code, I only learned to at all because I was cheap and/or poor cook/grad student that wanted to make little artsy musical things on the computer. I remember the first time I just downloaded pure data. No torrent, no cracks, it was just there for me and all it asked for was my patience.
The only reason I ever got into linux at all was because I ended up with some dinky acer chromebook for school but didn't want to stop making stuff. Crouton changed my life in a small way with that.
As I branched out and got more serious, learning web development, emacs, java, I never stopped feeling so irrationally lucky that it was all free, and always would be. Coming on here and other places to keep learning. It is to this day still the lovely forever hole I can excavate that costs only my sleep and electricity.
This is all not gone, but if I was just starting now, I'd find hn and so and coding twitter just like I did 10 years ago, but would be immediately turned off by this pervasive sense that "the way to do things now" is seemingly inseparable from a credit card number and monthly charge, however small. I just probably would not of gotten into it. It just wouldn't feel like its for me: "oh well I don't really know how to do this anyway, I can't justify spending money on it!" $0.76 for
50 loc is definitely nuts, but even $0.10 would of turned me way off. I had the same thoughts with all the web3 stuff too...
I know this speaks more to my money eccentricities than anything, and I know we dont really care on here about organic weirdo self teachers anymore (just productivity I guess). I am truly not even bemoaning the present situation, everyone has different priorities, and I am sure people are still having the exciting discovery of the computer like I did on their cursor ide or whatever. But I am personally just so so grateful the timeline lined up for me. I don't know if I'd have my passion for this stuff if I was born 10 years later than I was, or otherwise started learning now. But I guess we don't need the passion anymore anyway, its all been vectorized!
> But I am personally just so so grateful the timeline lined up for me.
I know the feeling. We still have access to the engineering thought processes responsible for some of the most amazing software feats ever accomplished (thru source repo history and mailing lists), just with access to the Internet. Of course there's a wealth of info available for free on the web for basically any profession, but for software engineering in particular it's almost direct access to world class teams/projects to learn from.
> but would be immediately turned off by this pervasive sense that "the way to do things now" is seemingly inseparable from a credit card number and monthly charge
To be effective you still need to understand and evaluate the quality of the output. There will always be a certain amount of time/effort required to get to that point (i.e., there's still no silver bullet).
> But I guess we don't need the passion anymore anyway, its all been vectorized!
We're not running out of things that can be improved. With or without these tools, the better you get, the more of the passion/energy that gets directed at higher levels of abstraction, i.e. thinking more about what to solve, tradeoffs in approaches, etc. instead of the minute details of specific solutions.
This doesn't make much sense to me. Is there some reason a kid today can't still learn to code? In the contrary, you have LLMs available that can answer your exact personalized questions. It's like having a free tutor. It's easier than it's ever been to learn for free.
I’m approaching middle age and have always wanted to learn to code and run servers, but would get caught up somewhere on tutorials and eventually give up in frustration.
Over the past year I have accomplished so much with the ever patient LLMs as my guide. It has been so much fun too. I imagine there are many others in my shoes and those that want to learn now have a much friendlier experience.
Yeah I'm middle aged and a competent programmer. But I hate learning new technologies these days. TypeScript was a huge hurdle to overcome. Working with ChatGPT made it so much more bearable. "Explain this to me. I need to do X, how? Why doesn't this work" etc.
This is definitely true in some ways. I was just talking around this point about spending money incrementally on aider, cursor, etc, and how it would have been a turnoff to me. But yes, all that I had back then people still have, and thats great.
Why learn if the computer can do it better than you and by the time you learn the roi on the market approaches 0? This wave of llm removed a lot of my interest in coding professionally
The sounds like the best way to get into coding. (For me it was wanting to realize game ideas to entertain myself.)
Money for a computer when I was getting into it was the credit-card part of it — there were no cheap Chromebooks then. (A student loan took care of the $1200 or so I needed for a Macintosh Plus.)
I suspect that's always the way of it though. There will be an easier way throwing money at a thing and there will be the "programming finds a way" way.
> this pervasive sense that "the way to do things now" is seemingly inseparable from a credit card number and monthly charge
…is true, but it only applies to experienced engineers who can sculpt the whole solution using these tools, not just random code. You need the whole learning effort to be able to ground the code the slop generators make. The passion absolutely helps here.
Note this is valid today. I have concerns that I’ll have different advice in 2027…
In 2028, the question will be who spent more money on lawsuits, and who spent more money on consultants to clean up their code base.
Jokes aside, code tools are best used in the hands of someone who is already trained and can verify bad code, and bad patterns at a glance.
AI code passes many tests. So does a lot of code written by us, for ourselves. When the code gets in front of users, especially the kind of genius users who learn how to fly by forgetting how to fall, then we learn many good habits.
In 2027 we'll have LLMs downloaded to our devices that are as good as Claude Code is today. (But as I have seen, as the leading edge of this stuff is always cooler than what you can run locally, we'll not be satisfied then with today's Claude Code.)
Tangential, but this reminds me of something someone said on Twitter that has resonated with me ever since. Startups targeting developers / building developer tooling are arguably one of the worst startups to build, because no matter how much of a steal the price is relative to the value you get, developers insist they can build their own or get by with an open-source competitor. We're as misguided on value as we are on efficiency and automation (more specifically, the old trope of a dev spending hours to automate something that takes minutes to do).
This is also why devs are not in charge of purchase decisions at tech companies. I don't mean FAANG but the usual tech shops.
Someone buys a tool and you have to use it. I think the startups selling dev tools are not selling to developers at all, but to the IT folks of these big tech firms.
Should they pull it off, it's not at all a bad startup to build. However, you need to now invest in a sales force that can sell to the Fortune 500. As a tech founder with no sales trope, this will be incredibly hard to pull off.
I digress, but yeah selling to devs is almost always a terrible idea since we all want to build our own stuff. That spirit may also be waning with the advent of Replit agent, Claude code and other tools.
> I think the startups selling dev tools are not selling to developers at all, but to the IT folks of these big tech firms.
They are often selling to IT managers against the advice of the developers and IT folks, and then they mostly don't get used because they don't actually add any value to the process.
I've noticed this tendency in myself and thought about the 'why' a lot, and I think it comes down to subconsciously factoring in the cost of lock-in, or worse, lack of access to fix/improve a tool I've come to rely on
For me, a larger part than "cost of lock-in" is the "hacker spirit", the curiosity to understand how it works.
Sure, I can pay google or fastmail to host a mailserver for me, but that deprives me of the joy of configuring and updating dovecot/postfix/etc, writing custom backup scripts, writing my own anti-spam tooling, etc. I want to know how all those pieces work.
Sure, I can pay kagi to search its index of webpages for me, but that deprives me of the joy of creating and running a botnet to index webpages, storing 100s of terrabytes of scraped data, and writing my own search code.
I think this spirit is totally lost on most people in this field. It’s tempting to say younger generations but it’s everyone. It always amazes me when I meet someone who has spent 10+ years in this field and doesn’t even care how anything but their shitty Kafka-Flink pipelines work.
If that; I’ve met plenty who only care that they work, not how they work.
As someone who works in infra and dabbles in coding, this is a continual bugbear, because often I’ll find an optimization while troubleshooting “my” problem, and the dev team is disinterested in implementing it. Their endpoints are meeting their SLO, so who cares?
I've honestly thought of hacker spirit as embodying a kind of homesteader ethos in a way. There's this homesteading book I bought a long time ago when I was in college, rich with illustrations on how to do everything from raise animals and grow food to building a house, processing lumber, drilling a well, everything. The same fascination I have with homesteading and DIY culture extends into my interest in technology, and I suspect this is the same with a lot of developers as well.
>We're as misguided on value as we are on efficiency and automation (more specifically, the old trope of a dev spending hours to automate something that takes minutes to do).
but automating something that takes minutes to do is Larry Wall's example of programmer laziness, and is a virtue.
of course - this needs obligatory conflicting XKCD comics
Isn't this a slippery slope into authoritarianism? How are people not more worried about this? The whole of checks and balances is this, this is democracy literally eroding away isn't it?
We’re in the authoritarianism part already, the 3 branches are aligned on this. The democrats whole campaign this election was to worry , but they are too committed to process and impartiality to have any effect (and let’s face it, most of congress is too old and/or too rich to give a damn what happens to our democracy in the first place)
Many of us made peace with this road a couple decades back and have been focused on building communities locally to survive it. Obama, Trump, Biden, GW ... they were all building the road to this moment together.
It's like watching one of those hydraulic press videos - the slow build up of the squeeze has been happening for awhile, we've finally arrived at the point where the pressure is really deforming stuff. But it was predictable for a long time.
Why can't they have better app moderation for apps going in in the first place? Prevention rather than cure? It almost feels like this is a mislead after they screwed up in the first place.
The title is wrong; prevention is what they do. The actual quote from the article is:
> As a result, we prevented 2.36 million policy-violating apps from being published on Google Play and banned more than 158,000 bad developer accounts that attempted to publish harmful apps.
Reminds me of Vault-Tec from the Fallout universe trying to "win the capitalism game" and "optimize shareholder value" after a nuclear apocalypse, going so far as to facilitate nuclear war in order to raise their value as a "defense company"
Location: Canada / New York
Remote: Remote OK
Willing to relocate: Maybe
Technologies: Python, Django, Javascript, iOS, ML, LLMs, Distributed Systems
Résumé/CV: https://www.linkedin.com/in/personjerry/
Email: Click my username
Previously at Facebook, Dendron (YC W21). Engineering leader with 9+ years of experience at both FAANG scale and at startup intensity. I also still build systems and touch code every day so I can IC as well. Primarily looking for opportunities in NYC or remote US, but there's many cities I love. Currently in some conversations with big tech and considering competitive options :)
Ah yes, they just don't make cyanide ice cream like grandma used to