291–300 of 958 posts
Earlier quoted context omitted.
> 2000 lines is 10x more time consuming to review than 200 Very far from the truth in practice, every line of code isn't as difficult/easy to review as the other.
Holy shit, read the words I wrote. Ceteris Paribus. Assume the 200 lines and 2000 lines have a similar distribution of complexity.
Earlier quoted context omitted.
It's about responsibility. If I get pwned because my AI agent wrote code that had a security vulnerability, none of my users are going to accept the excuse that I used AI and it's a brave new world. I will get the blame, not Anthropic or OpenAI or Google but me. The same goes for if my AI generated code leads to data loss, or downtime, or if uses too many resources, or it doesn't scale, or it gives out error messages…
On the other hand, I don’t need to review carefully every line of code in my thumbnail generator and associated UI. My nonexistent backend isn’t going to be pwned if there is a bug in the thumbnail generation. After the QA testing on my device, a quick scroll through of the code is enough. Maybe prompt „are errors during thumbnail generation caught to prevent app crashes?“ if we‘re feeling extra cautious today. And j…
Hmm. Historically image editing was one of the easier to exploit security holes in many systems. How do you feel about having unknown entities having shell inside your datacenter or vpc?
Perhaps I've missed a few weeks worth of progress, but I don't think that AIs have become more trustworthy, the errors are just more subtle. If the code doesn't compile, that's easy to spot. If the code compiles but doesn't work, that's still somewhat easy to spot. If the code compiles and works, but it does the wrong thing in some edge case, or has a security vulnerability, or introduces tech debt or dubious archite…
Earlier quoted context omitted.
"many engineers have not yet tried things that became feasible over the last months" I have heard this statement every single day for 2 years and yet we still have no companies compressing 10 years into 1 year thus exploding past all the incumbents who don't "get it".
If you want an example of a project that condensed 5 years into 6 months and exploded past the competition I suggest looking at OpenClaw. The first line of code was written on November 25th. It achieved adoption in the "personal agents" space that far exceeded the other companies that had tried the same thing. (Whether or not you trust the quality of the software you can't deny the impact it had in such a short time.…
Earlier quoted context omitted.
This has generally been the case, though. As mentioned in the post, "You want solutions that are proven to work before you take a risk on them" remains true and will be place where the edges are found.
It's about responsibility. If I get pwned because my AI agent wrote code that had a security vulnerability, none of my users are going to accept the excuse that I used AI and it's a brave new world. I will get the blame, not Anthropic or OpenAI or Google but me. The same goes for if my AI generated code leads to data loss, or downtime, or if uses too many resources, or it doesn't scale, or it gives out error messages…
First of all, building a system that constrains the output of the AI sufficiently, whether that's typing, testing, external validation, or manual human review in extremis. That gets you the best result out of whatever harness or orchestration you're using.
Secondly, there's the level at which you're intervening, something along the hierarchy of "validate only usage from the customer perspective" to "review, edit, and validate every jot and tiddle of the codebase and environment". I think for relatively low importance things reviewing at the feature level (all code, but not interim diffs) is fine, but if you're doing network protocol you better at least validate everything carefully with fuzzing and prop testing or something like that.
And then you've got how you structure your feedback to the LLM itself - is it an in-the-loop chat process, an edit-and-retry spec loop, go-nogo on a feature branch, or what? How does the process improve itself, basically?
I agree with you entirely that the responsibility rests on the human, but there are a variety of ways to use these things that can increase or decrease the quality of code to time spent reviewing, and obviously different tasks have different levels of review scrutiny, as well.
Earlier quoted context omitted.
"many engineers have not yet tried things that became feasible over the last months" I have heard this statement every single day for 2 years and yet we still have no companies compressing 10 years into 1 year thus exploding past all the incumbents who don't "get it".
If you want an example of a project that condensed 5 years into 6 months and exploded past the competition I suggest looking at OpenClaw. The first line of code was written on November 25th. It achieved adoption in the "personal agents" space that far exceeded the other companies that had tried the same thing. (Whether or not you trust the quality of the software you can't deny the impact it had in such a short time.…
Which is exactly why you can't use it as an example, there is no control. This is basic stuff.
The scary part is that codebases are getting layers of AI complexity, that it's going to cost $$$ to have the latest model decipher and make changes as no human can understand the code anymore. Pretty soon there is no code reuse and we're burning money reinventing the wheel over and over.
Earlier quoted context omitted.
If you want an example of a project that condensed 5 years into 6 months and exploded past the competition I suggest looking at OpenClaw. The first line of code was written on November 25th. It achieved adoption in the "personal agents" space that far exceeded the other companies that had tried the same thing. (Whether or not you trust the quality of the software you can't deny the impact it had in such a short time.…
OpenClaw is definitely not a "5 years" project pre-AI though. That was more like a month of greenfield work compressed into a weekend -- which is still really impressive, don't get me wrong! -- but I think the point is we're not seeing mature , legacy codebases get outcompeted by new, agile, AI-driven codebases; we're seeing greenfield projects get spun up faster. Which, again, is still impressive and valuable. If ag…
If that were true, all of these anti-AI greybeards who have been in the game for 30 years would all own their own jets.