Model Comparison
Model Editorial Structural Class Conf SETL Theme
@cf/meta/llama-3.3-70b-instruct-fp8-fast lite ND ND 0.83
@cf/meta/llama-3.3-70b-instruct-fp8-fast lite 0.00 -0.13 Neutral 0.80 0.13 Tech productivity
@cf/meta/llama-4-scout-17b-16e-instruct lite ND ND 0.80
@cf/meta/llama-4-scout-17b-16e-instruct lite +0.10 -0.13 Neutral 0.80 0.17 AI and Productivity
claude-haiku-4-5-20251001 +0.16 +0.11 Mild positive 0.46 0.17 Labor Rights & Meaningful Work
Section @cf/meta/llama-3.3-70b-instruct-fp8-fast lite @cf/meta/llama-3.3-70b-instruct-fp8-fast lite @cf/meta/llama-4-scout-17b-16e-instruct lite @cf/meta/llama-4-scout-17b-16e-instruct lite claude-haiku-4-5-20251001
Preamble ND ND ND ND 0.15
Article 1 ND ND ND ND 0.20
Article 2 ND ND ND ND -0.05
Article 3 ND ND ND ND 0.10
Article 4 ND ND ND ND ND
Article 5 ND ND ND ND ND
Article 6 ND ND ND ND 0.25
Article 7 ND ND ND ND 0.20
Article 8 ND ND ND ND ND
Article 9 ND ND ND ND ND
Article 10 ND ND ND ND 0.15
Article 11 ND ND ND ND ND
Article 12 ND ND ND ND -0.10
Article 13 ND ND ND ND 0.10
Article 14 ND ND ND ND 0.14
Article 15 ND ND ND ND 0.13
Article 16 ND ND ND ND ND
Article 17 ND ND ND ND 0.05
Article 18 ND ND ND ND 0.24
Article 19 ND ND ND ND 0.44
Article 20 ND ND ND ND 0.10
Article 21 ND ND ND ND 0.05
Article 22 ND ND ND ND 0.19
Article 23 ND ND ND ND 0.14
Article 24 ND ND ND ND 0.10
Article 25 ND ND ND ND 0.20
Article 26 ND ND ND ND 0.34
Article 27 ND ND ND ND 0.19
Article 28 ND ND ND ND 0.13
Article 29 ND ND ND ND 0.20
Article 30 ND ND ND ND ND
+0.16 Codegen Is Not Productivity (www.antifound.com S:+0.11 )
81 points by donutshop 7 days ago | 100 comments on HN | Mild positive Moderate agreement (3 models) Editorial · v3.7 · 2026-03-15 22:37:56 0
Summary Labor Rights & Meaningful Work Advocates
This blog post advocates for recognizing that software development productivity is fundamentally about intellectual work, human collaboration, and long-term system sustainability—not code generation volume. The author challenges mainstream LLM productivity narratives by grounding critique in software engineering principles, emphasizing rights to meaningful work, fair conditions, community participation, and worker accountability for production systems. The piece demonstrates strong commitment to free expression, education, and worker agency in technological decisions.
Rights Tensions 3 pairs
Art 23 Art 19 Right to meaningful work and human judgment conflicts with pressure toward automated code generation; content resolves this by asserting human agency and explicit choice in technology adoption.
Art 22 Art 19 Social security for maintenance workers and operational stability conflicts with rapid code generation pushing costs forward; content resolves by advocating sustainable pace and reduced code volume.
Art 12 Art 23 Privacy and data handling understanding (requiring human comprehension of code) conflicts with developer ability to perform meaningful work when systems become unmaintainable; content resolves by emphasizing human understanding requirement.
Article Heatmap
Preamble: +0.15 — Preamble P Article 1: +0.20 — Freedom, Equality, Brotherhood 1 Article 2: -0.05 — Non-Discrimination 2 Article 3: +0.10 — Life, Liberty, Security 3 Article 4: ND — No Slavery Article 4: No Data — No Slavery 4 Article 5: ND — No Torture Article 5: No Data — No Torture 5 Article 6: +0.25 — Legal Personhood 6 Article 7: +0.20 — Equality Before Law 7 Article 8: ND — Right to Remedy Article 8: No Data — Right to Remedy 8 Article 9: ND — No Arbitrary Detention Article 9: No Data — No Arbitrary Detention 9 Article 10: +0.15 — Fair Hearing 10 Article 11: ND — Presumption of Innocence Article 11: No Data — Presumption of Innocence 11 Article 12: -0.10 — Privacy 12 Article 13: +0.10 — Freedom of Movement 13 Article 14: +0.14 — Asylum 14 Article 15: +0.13 — Nationality 15 Article 16: ND — Marriage & Family Article 16: No Data — Marriage & Family 16 Article 17: +0.05 — Property 17 Article 18: +0.24 — Freedom of Thought 18 Article 19: +0.44 — Freedom of Expression 19 Article 20: +0.10 — Assembly & Association 20 Article 21: +0.05 — Political Participation 21 Article 22: +0.19 — Social Security 22 Article 23: +0.14 — Work & Equal Pay 23 Article 24: +0.10 — Rest & Leisure 24 Article 25: +0.20 — Standard of Living 25 Article 26: +0.34 — Education 26 Article 27: +0.19 — Cultural Participation 27 Article 28: +0.13 — Social & International Order 28 Article 29: +0.20 — Duties to Community 29 Article 30: ND — No Destruction of Rights Article 30: No Data — No Destruction of Rights 30
Negative Neutral Positive No Data
Aggregates
E
+0.16
S
+0.11
Weighted Mean +0.17 Unweighted Mean +0.15
Max +0.44 Article 19 Min -0.10 Article 12
Signal 24 No Data 7
Volatility 0.11 (Medium)
Negative 2 Channels E: 0.6 S: 0.4
SETL +0.17 Editorial-dominant
FW Ratio 58% 70 facts · 50 inferences
Agreement Moderate 3 models · spread ±0.111
Evidence 46% coverage
4H 16M 4L 7 ND
Theme Radar
Foundation Security Legal Privacy & Movement Personal Expression Economic & Social Cultural Order & Duties Foundation: 0.10 (3 articles) Security: 0.10 (1 articles) Legal: 0.20 (3 articles) Privacy & Movement: 0.07 (4 articles) Personal: 0.14 (2 articles) Expression: 0.20 (3 articles) Economic & Social: 0.16 (4 articles) Cultural: 0.27 (2 articles) Order & Duties: 0.17 (2 articles)
HN Discussion 19 top-level · 27 replies
nyrulez 2026-03-15 16:41 UTC link
Bold claims that writing code was never the bottleneck. It may not be the only bottleneck but we conveniently move goal posts now that there is a more convenient mechanism and our profession is under threat.
emp17344 2026-03-15 16:47 UTC link
The collaboration aspect is what many AI enthusiasts miss. As humans, our success is dependent on our ability to collaborate with others. You may believe that AI could replace many individual software engineers, but if it does so at the expense of harming collaboration, it’s a massive loss. AI tools are simply not good at collaborating. When you add many humans to a project, the result becomes greater than the sum of its parts. When you add many AI tools to a project, it quickly becomes a muddled mess.
gaigalas 2026-03-15 16:58 UTC link
In practical terms, "productivity" is any metric that people with power can manipulate (cheating numbers, changing narratives, etc) to affect behavior of others to their interests.

ALL OF IT is meaningless. It's a pointless discussion.

autonomousErwin 2026-03-15 16:59 UTC link
I think there's some goldilocks speed limit for using these tools relative to your skillset. When you're building, you forget that you're also learning - which is why I actually favour some AI code editors that aren't as powerful because it gets me to stop and think.
demorro 2026-03-15 17:03 UTC link
A well considered article, despite the author categorizing it as a rant. I appreciate the appendix quotations, as well as the acknowledgement that they are appeals to authority.

Whilst the author clearly has a belief that falls down on one side of the debate, I hope folks can engage with the "Should we abandon everything we know" question, which I think is the crux of things. Evidence that AI-driven-development is a valuable paradigm shift is thin on the ground, and we've done paradigm shifts before which did not really work out, despite massive support for them at the time. (Object-Oriented-Everything, Scrum, etc.)

jwilliams 2026-03-15 17:03 UTC link
> Humans and LLMs both share a fundamental limitation. Humans have a working memory, and LLMs have a context limit.

But there’s a more important difference: I can’t spin up 20 decent human programmers from my terminal.

The argument that "code was never the bottleneck" is genuinely appealing, but it hasn’t matched my experience at all. I’m getting through dramatically more work now. This is true for my colleagues too.

My non-technical niece recently built a pretty solid niche app with AI tools. That would have been inconceivable a few years ago.

avabuildsdata 2026-03-15 17:04 UTC link
honestly the thing that trips me up is when codegen makes me feel productive but I haven't actually validated anything. like I'll have claude write a whole data pipeline in 20 minutes and then spend 2 hours debugging edge cases it didn't think about because it doesn't know our data

the speed is real but it mostly just moves where I spend my time. less typing, more reading and testing. which is... fine? but it's not the 10x thing people keep claiming

nubg 2026-03-15 17:07 UTC link
For me it's simple:

1. Assume you're to work on product/feature X.

2. If God were to descend and give you a very good, reality-tested spec:

3. Would you be done faster? Of course, because as every AI doomer says, writing code was never the bottleneck!!1!

4. So the only bottleneck is getting to the spec.

5. Guess what AI can help you with as well, because you can iterate out multiple versions with little mental effort and no emotional sunk cost investment?

ergo coding is a solved problem

jwpapi 2026-03-15 17:13 UTC link
I have to be honest. I’ve written a lot of pro-ai / dark-software articles and I think Im due an update, cause it worked great, till it didn’t.

I could write a lot about what I’ve tried and learnt, but so far this article is a very based view and matches my experience.

I definitely suffered under the unnecessary complexity and wished to never’ve used AI at moments and even with OPUS 4.6 I could feel how it was confused and couldn’t understand business objectives really. It became way faster to jump in code, clean it up and fix it myself. I’m not sure yet where and how the line is and where it will be.

zer00eyz 2026-03-15 17:15 UTC link
I went to look at some of the authors other posts and found this:

https://www.antifound.com/posts/advent-of-code-2022/

So much of our industry has spent the last two decades honing itself into a temple built around the idea of "leet code". From the interview to things like advent of code.

Solving brain teasers, knowing your algorithms cold in an interview was always a terrible idea. And the sort of engineers it invited to the table the kinds of thinking it propagated were bad for our industry as a whole.

LLM's make this sort of knowledge, moot.

The complaints about LLM's that lack any information about the domains being worked in, the means of integration (deep in your IDE vs cut and paste into vim) and what your asking it to do (in a very literal sense) are the critical factors that remain "un aired" in these sorts of laments.

It's just hubris. The question not being asked is "Why are you getting better results than me, am I doing something wrong?"

jwpapi 2026-03-15 17:16 UTC link
There is a saying you need to write an essay 3 times. The first time its puked out, the second is decent and the third is good.

It’s quite similar with code, and with code less is more. for try 1 and 2

greggyb 2026-03-15 17:17 UTC link
Hey, author here. Never thought I'd see my pokey little blog on HN and all that.

Happy to discuss further.

swalsh 2026-03-15 17:17 UTC link
Speak for yourself, I have never thrown away code at this rate in my entire career. I couldn't keep up this pace without AI codegen.
vinceguidry 2026-03-15 17:18 UTC link
I recently started using AI for personal projects, and I find it works really well for 'spike' type tasks, where what you're trying to do is grow your knowledge about a particular domain. It's less good at discovering the correct way of doing things once you've decided on a path forward, but still more useful than combing through API docs and manpages yourself.

It might not actually deliver working things all that much faster than I could, but I don't feel mentally drained by the process either. I used to spend a lot of time reading architecture docs in order to understand available solutions, now I can usually get a sense for what I need to know just from asking ChatGPT how certain things might be done using X tool.

In the last few days, I've stood up syncthing, tailscale with a headscale control plane, and started making working indicators and strategies in PineScript, TradingView's automated trading platform. Things I had no energy for or would have been weeklong projects take hours or a day or so. AI's strengths synergize really well with how humans want to think.

I just paste an error message in, and ChatGPT figures out what I'm trying to do from context, then gives me not just a possible resolution, but also why the error is happening. The latter is just as useful as the former. It's wrong a lot, but it's easy to suss out.

ChicagoDave 2026-03-15 17:44 UTC link
I continue to jump into these discussions because I feel like these upvoted posts completely miss what’s happening…

- guardrails are required to generate useful results from GenAI. This should include clear instructions on design patterns, testing depth, and iterative assessments.

- architecture decision records are one useful way to prevent GenAI from being overly positive.

- very large portions of code can be completely regenerated quickly when scope and requirements change. (skip debugging - just regenerate the whole thing with updated criteria)

- GenAI can write thorough functional and behavioral unit tests. This is no longer a weakness.

- You must suffer the questions and approvals. At no time can you let agents run for extended periods of time on progressive sets of work. You must watch what is generated. One thing that concerns me about the new 1mm context on Claude Code is many will double down on agent freedom. You can’t. You must watch the results and examine functionality regularly.

- No one should care about actual code ever again. It’s ephemeral. The role of software engineering is now molding features and requirements into functional results. Choosing Rust, C#, Java, or Typescript might matter depending on the domain, but then you stop caring and focus on measuring success.

My experience is rolled up in https://devarch.ai/ and I know I get productive and testable results using it everyday on multiple projects.

eleventhborn 2026-03-15 17:49 UTC link
I feel there is a set of codebases in which LLMs aren't showing the 2-10x lift in productivity.

There is also a set of codebases in which LLMs are one-shotting the most correct code and even finding edgecases that would've been hard to find in human reviews.

At a surface level, it seems obvious that legacy codebases tend to fall in the first category and more greenfield work falls in the second category.

Perhaps, this signals an area of study where we make codebases more LLM-friendly. It needs more research and a catchy name.

Also, certain things that we worry about as software artisans like abstractions, reducing repeated code, naming conventions, argument ordering,... is not a concern for LLMs. As long as LLMs are consistent in how they write code.

For e.g. One was taught that it is bad to have multiple "foo()" implementations. In LLM world, it isn't _that_ bad. You can instruct the LLM to "add feature x and fix all the affected tests" (or even better "add feature x to all foo()") and if feature x relies on "foo()", it fixes every foo() method. This is a big deal.

slopinthebag 2026-03-15 17:58 UTC link
It's so difficult to quantify productivity over an entire field, especially when it's so vast. Chris Lattner recently concluded this about LLM tooling [0]:

> AI systems can internalize the textbook knowledge of a field and apply it coherently at scale. AI can now reliably operate within established engineering practice. This is a genuine milestone that removes much of the drudgery of repetition and allows engineers to start closer to the state of the art.

This matches my experience, there is a lot of code that we probably should not need to write and rewrite anymore but still do because this field has largely failed at deriving complete and reusable solutions to trivial problems - there is a massive coordination problem that has fragmented software across the stack and LLMs provide one way of solving it by generating some of the glue and otherwise trivial but expensive and unproductive interop code required.

But the thing about productivity is that it's not one thing and cannot be reduced to an anecdote about a side-project, or a story about how a single company is introducing (or mandating) AI tooling, or any single thing. Being able to generate a bunch of code of varying quality and reliability is undeniably useful, but there are simply too many factors involved to make broad sweeping claims about an entire industry based on a tool that is essentially autocomplete on crack. Thus it's not surprising that recent studies have not validated the current hype cycle.

[0] https://www.modular.com/blog/the-claude-c-compiler-what-it-r...

galbar 2026-03-15 18:02 UTC link
This article describes the body of knowledge I was taught when I joined the industry and parallels with my experience with and thoughts about AI.

I have come to the realization that most people in the industry don't know this body of knowledge, or even that it exists.

I'm now seeing the same people trying to solve their ineffectiveness with AI.

I don't know what to think about this situation. My intuition hints at it not being good.

fulafel 2026-03-15 20:18 UTC link
Productivity in econ means how many units of output you get, not if they are good. So codegen is productivity in this sens, but not what you want.
simianwords 2026-03-15 16:55 UTC link
AI will allow us to collaborate on higher level decisions and not on whether we should use for loops or functional interfaces.
Aurornis 2026-03-15 16:56 UTC link
> AI tools are simply not good at collaborating

My primary use of LLM tools is as a collaborator.

I agree that if you try to use the LLM as a wholesale outsourcing of your thought process the results don’t scale. That’s not the only way to use them, though.

demorro 2026-03-15 16:56 UTC link
There's plenty of evidence of this line of thinking even from before the turn of the Millennium. Mythical Man Month, No Silver Bullet, Code Complete, they all gesture at this point.
orphea 2026-03-15 16:57 UTC link

  our profession is under threat.
It is. But I don't think it's AI that threatens it. It's susceptible to hype people who, unfortunately, have the power over people's jobs. C-level management who don't know anything better than parroting what others in the industry are saying. How is that "all engineers will be replaced in 6 months" going?
Frieren 2026-03-15 17:01 UTC link
This is a case of "depends on the project".

For very small projects, code may be the main bottleneck. Just to write the code is what takes most of the time. Adding code faster can accelerate development.

For larger projects, design, integration, testing, feature discovery, architecture, bug fixing, etc. takes most of the time. Adding code faster may slow down development and create conflicts between teams.

Discussing without a common context makes no sense in this situation.

So, depending on your industry and the size of the projects that you have worked on one thing or the other may be true.

Sparkle-san 2026-03-15 17:04 UTC link
Writing good code might be a bottleneck and the same can't be said about code in general.
hibikir 2026-03-15 17:05 UTC link
I look at it backwards: A few humans improves a project. But once you get to sufficient sizes, principal-agent problems dominate. What is good for a division and what is good for the company disagree. What is good for a developer that needs a big project for their promotion package is not what the company needs. A company with a headcount of 700 is more limber and better aligned than when it's 3,000 or 30,000. It's amazing how little alignment there ever is when you get to the 300k range.

AI, if anything, is amazing at collaborating. It's not perfectly aligned, but you sure can get it to tell you when your idea is unsound, all while having lessened principal-agent issues. Anything we can do to minimize the number of people that need to align towards a goal, the more effectively we can build, precisely due to the difficulties of marshalling large numbers of people. If a team of 4 can do the same as a team of 10, you should always pick the team of 4, even if they are more expensive put together than the 10.

demorro 2026-03-15 17:07 UTC link
Would you entertain the idea that "work was never the bottleneck", or even "building products was never the bottleneck"?

We need to address Jevons' Paradox somehow.

nubg 2026-03-15 17:08 UTC link
Would getting to the same edge-case-free outcome have taken you less than 2h20min if you didn't have AI?

I think it would typically have taken you longer.

felipellrocha 2026-03-15 17:10 UTC link
I guess that what people debate on here is what “decent” mean. From my experience, these llms spit out dog shit code, so 20 agents equal 20x more dog shit.
dolebirchwood 2026-03-15 17:14 UTC link
> When you add many humans to a project, the result becomes greater than the sum of its parts. When you add many AI tools to a project, it quickly becomes a muddled mess.

I have absolutely been on projects where there were too many cooks in the kitchen, and adding more people to the team only led to additional chaos, confusion, and complexity. Ever been in a meeting where a designer, head of marketing, and the CTO are all giving feedback on what size font a button should be? I certainly have, and it's absurd.

One of my worst experiences arose due to having a completely incompetent PM. Absolutely no technical knowledge; couldn't even figure out how to copy and paste a URL if his life depended on it. He eventually had to be be removed from a major project I was on, and I was asked to take over PM duties, while also doing my dev work. I was actually happy to do so, because I was already having to spend hours babysitting him; now I could just get the same tasks done without the political BS.

Could adding many AI tools to a project become problematic? Maybe. But let's not pretend throwing more humans at a project is going to lead to some synergistic promised land.

greggyb 2026-03-15 17:18 UTC link
Unfortunately, this post was published at the puked out phase (;

(author here)

greggyb 2026-03-15 17:23 UTC link
I actually consider that the claim is not that bold, and in fact has been common in our industry for most of the short time it has been around. I included a few articles and studies with time breakdowns of developer activity that I think help to illustrate this.

If an activity (getting code into source files) used to take up <50% of the time of programmers, then removing that bottleneck cannot even double the throughput of the process. This is not taking into account non-programmer roles involved in software development. This is akin to Amdahl's law when we talk about the benefits of parallelism.

I made no argument with regard to threat to the profession, and I make none here.

greggyb 2026-03-15 17:27 UTC link
I didn't set out to teach you anything, change your behavior, or give you practical takeaways, so it's a rant (: Emotions can be expressed with citations.

I am fully on board with gen AI representing a paradigm shift in software development. I tried to be careful not to take a stance on other debates in the larger conversation. I just saw too many people talking about how much code they're generating as proof statements when discussing LLMs. I think that, specifically---i.e., using LOC generated as the basis of any meaningful argument about effectiveness or productivity---is a silly thing to do. There are plenty of other things we should discuss besides LOC.

demorro 2026-03-15 17:28 UTC link
Hey, I like your writing. You got an rss feed or anything?
greggyb 2026-03-15 17:30 UTC link
> The complaints about LLM's that lack any information about the domains being worked in, the means of integration (deep in your IDE vs cut and paste into vim) and what your asking it to do (in a very literal sense) are the critical factors that remain "un aired" in these sorts of laments.

I'm not sure if this is a direct response to the article or a general point. The article includes an appendix about my use of LLMs and the domains I have used them in.

greggyb 2026-03-15 17:38 UTC link
I'd recommend you read the book referenced in the conclusion: https://link.springer.com/chapter/10.1007/978-1-4842-4221-6_...

The full PDF is available for download. It's mostly a series of essays, so you can pick and choose and read nonlinearly. It's worth thinking about beyond nihilistic takes.

foolserrandboy 2026-03-15 17:40 UTC link
XD
greggyb 2026-03-15 17:47 UTC link
The post is about using LOC as a metric when making any sort of point about AI. Nowhere do I suggest someone shouldn't use it, nor that they should expect negative results if they opt to.
sarchertech 2026-03-15 17:47 UTC link
Did you read the article? I don’t think that refutes anything the author said even a little bit.
ip26 2026-03-15 17:54 UTC link
No one should care about actual code ever again. It’s ephemeral.

Caveat: it still works best in a codebase that is already good. So while any one line of code is ephemeral, how is the overall codebase trending? Towards a bramble, or towards a bonsai?

If the software is small and not mission critical, it doesn’t matter if it becomes a bramble, but not all software is like that.

slopinthebag 2026-03-15 18:02 UTC link
> - No one should care about actual code ever again. It’s ephemeral. The role of software engineering is now molding features and requirements into functional results. Choosing Rust, C#, Java, or Typescript might matter depending on the domain, but then you stop caring and focus on measuring success.

I think this has always been the case. "Bad programmers worry about the code. Good programmers worry about data structures and their relationships." Perhaps you mean that they shouldn't worry about structures & relationships either but I think that is a fools errand. Although to be fair neither of those need to be codified in the code itself, but ignore those at your own peril...

felipellrocha 2026-03-15 18:05 UTC link
Man, if this were true we’d see a crazy, massive explosion of quality products being written, and launched. While we see some use, i don’t perceive an acceleration. In fact, i see a lot of trivial bugs being deployed to prod.
slopinthebag 2026-03-15 18:08 UTC link
4 doesn't follow from 3
sarchertech 2026-03-15 18:12 UTC link
And then it turns out God wrote the spec in code because that’s what any spec sufficient to produce the same program from 2 different teams/LLMs would be.
sarchertech 2026-03-15 19:06 UTC link
> No one should care about actual code ever again. It’s ephemeral.

> very large portions of code can be completely regenerated quickly when scope and requirements change.

This is complete and utter nonsense coming from someone who isn't actually sticking around maintaining a product long enough in this manner to see the end result of this.

All of this advice sounds like it comes from experience instead of theoretical underpinning or reasoning from first principles. But this type of coding is barely a year old, so there's no way you could have enough experience to make these proclamations.

Based on what I can talk about from decades of experience and study:

No natural language specification or test suite is complete enough to allow you to regenerate very large swaths of code without changing thousands of observable behaviors that will be surfaced to users as churn, jank, and broken workflows. The code is the spec. Any spec detailed enough to allow 2 different teams (or 2 different models or prompts) to produce semantically equivalent output is going to be functionally equivalent to code. We as an industry have learned this lesson multiple times.

I'd bet $1,000 that there is no non-trivial commercial software in existence where you could randomly change 5% of the implementation while still keeping to the spec and it wouldn't result in a flood of bug reports.

The advantage of prompting in a natural language is that the AI fills in the gaps for you. It does this by making thousands of small decisions when implementing your prompt. That's fine for one offs, and it's fine if you take the time to understand what those decisions are. You can't just let the LLM change all of those decision on a whim, which is the natural result of generating large swaths of code, ignoring it, and pretending it's ephemeral.

antonvs 2026-03-15 19:32 UTC link
My career predates the leetcode phenomenon, and I always found it mystifying. My hot take is that it’s what happens when you’re hiring what are essentially human compilers: they can spit out boilerplate solutions at high speed, and that’s what leetcode is testing for.

For someone like that, LLMs are much closer to literally replacing what they do, which seems to explain a lot of the complaints. They’re also not used to working at a higher level, so effective LLM use doesn’t come naturally to them.

Editorial Channel
What the content says
+0.35
Article 19 Freedom of Expression
High Advocacy Practice
Editorial
+0.35
SETL
+0.23

Content directly advocates for free expression and information sharing. Publishes detailed technical argument contrary to mainstream LLM enthusiasm. Shares knowledge through appendices, references, and peer recommendations.

+0.30
Article 18 Freedom of Thought
High Advocacy Practice
Editorial
+0.30
SETL
+0.21

Content strongly advocates for freedom of thought and expression in software development. Argues developers must explicitly consider their own positions on LLM adoption rather than accepting mainstream narrative. Critiques implicit pressures toward technological conformity.

+0.30
Article 26 Education
High Advocacy Practice
Editorial
+0.30
SETL
+0.21

Content strongly advocates for right to education and access to knowledge. Extensive citations, references, and sharing of foundational materials. Teaches critical thinking about productivity metrics and software engineering principles.

+0.25
Article 6 Legal Personhood
Medium Framing
Editorial
+0.25
SETL
ND

Content affirms right to recognition as a person; critiques LLM-centric narratives that diminish human programmer agency, reasoning, and recognition. Emphasizes that responsibility and credit remain with human practitioners.

+0.25
Article 22 Social Security
Medium Framing Practice
Editorial
+0.25
SETL
+0.19

Content advocates for social security and welfare through emphasis on sustainable development practices. Critiques LLM-driven development as creating maintenance burden that shifts costs to future workers and operations teams.

+0.25
Article 27 Cultural Participation
High Advocacy Practice
Editorial
+0.25
SETL
+0.19

Content advocates for participation in cultural life of software community. Emphasizes shared understanding, code as medium for human expression, and collective knowledge. Critiques automation that diminishes cultural participation.

+0.20
Article 1 Freedom, Equality, Brotherhood
Medium Framing
Editorial
+0.20
SETL
ND

Content emphasizes equal reasoning capacity of humans and systems, argues against automation that undermines human judgment in software decisions. Advocates for human responsibility and deliberate choice in development practices.

+0.20
Article 7 Equality Before Law
Medium Framing
Editorial
+0.20
SETL
ND

Content advocates for equal treatment in responsibility and code review. Argues that all developers, regardless of tool use, must maintain equal standards. Emphasizes peer review as fundamental practice.

+0.20
Article 14 Asylum
Medium Framing Practice
Editorial
+0.20
SETL
+0.17

Content affirms right to seek asylum and protection; implicitly through framing of developer autonomy and freedom to choose development practices without coercion toward LLM dependency. Advocates for retention of human control over technological direction.

+0.20
Article 23 Work & Equal Pay
Medium Framing Practice
Editorial
+0.20
SETL
+0.17

Content advocates for right to work with fair conditions and protections. Emphasizes programmer autonomy, meaningful work, and protection against deskilling through automation. Advocates for work that requires human judgment.

+0.20
Article 25 Standard of Living
Medium Framing
Editorial
+0.20
SETL
ND

Content advocates for standard of living through emphasis on product quality, user welfare, and sustainable development. Critiques practices that compromise system reliability and user experience.

+0.20
Article 29 Duties to Community
Medium Framing
Editorial
+0.20
SETL
ND

Content advocates for duties and responsibilities alongside rights; repeatedly emphasizes programmer accountability, peer responsibility, and community obligation.

+0.15
Preamble Preamble
Medium Framing
Editorial
+0.15
SETL
ND

Content frames programming as intellectual work requiring human reasoning and collaboration, values dignity of human labor in software development, implicitly affirms human agency in work and decision-making.

+0.15
Article 10 Fair Hearing
Medium Framing
Editorial
+0.15
SETL
ND

Content implicitly affirms fair hearing and due process in development practices; emphasizes transparent code review, explicit standards, and peer accountability. Critiques automation that bypasses human judgment in decisions affecting system behavior.

+0.15
Article 15 Nationality
Medium Framing Practice
Editorial
+0.15
SETL
+0.09

Content affirms right to nationality and belonging to community; emphasizes programmer community norms and responsibility to collective software ecosystem. Critiques LLM practices that undermine community sustainability.

+0.15
Article 28 Social & International Order
Medium Framing Practice
Editorial
+0.15
SETL
+0.09

Content advocates for social and international order supporting realization of rights; emphasizes software as infrastructure affecting multiple stakeholders, not just individual developers.

+0.10
Article 3 Life, Liberty, Security
Medium Framing
Editorial
+0.10
SETL
ND

Content emphasizes right to meaningful work and security; critiques LLM productivity metrics that prioritize speed over correctness, stability, and sustainable practices. Argues that well-functioning software (security and reliability) requires human care.

+0.10
Article 13 Freedom of Movement
Low Framing
Editorial
+0.10
SETL
ND

Content tangentially addresses freedom of movement and residence through critique of how LLM-driven development may create brittle, unmaintainable systems that constrain users' ability to migrate or modify software.

+0.10
Article 20 Assembly & Association
Medium Framing
Editorial
+0.10
SETL
ND

Content advocates for peaceful assembly and association in software development communities. Emphasizes community standards, peer review, and collective responsibility against isolation of individual productivity.

+0.10
Article 24 Rest & Leisure
Medium Framing
Editorial
+0.10
SETL
ND

Content advocates for rest, leisure, and reasonable working hours through critique of burnout-inducing acceleration. Emphasizes deliberate pacing of work and sustainability.

+0.05
Article 17 Property
Low Framing
Editorial
+0.05
SETL
ND

Content implicitly addresses property rights; argues against uncritical appropriation of LLM outputs and advocates for using existing libraries and established patterns rather than generating custom code, respecting established intellectual property.

+0.05
Article 21 Political Participation
Low Framing
Editorial
+0.05
SETL
ND

Content tangentially addresses participation in governance through emphasis on explicit choice and deliberate consideration of technological direction. Advocates against implicit adoption of practices without community deliberation.

-0.05
Article 2 Non-Discrimination
Low
Editorial
-0.05
SETL
ND

Content does not directly address non-discrimination, though implicit in critique is concern that LLM productivity claims may be equally applied across different development contexts without acknowledging structural differences.

-0.10
Article 12 Privacy
Medium Framing
Editorial
-0.10
SETL
ND

Content raises concerns about loss of privacy in rapid LLM-driven development; customers exposed to systems developed without full human understanding. Concerns about personal data handling when developers cannot fully explain system behavior.

ND
Article 4 No Slavery

No discussion of slavery or forced labor.

ND
Article 5 No Torture

No discussion of torture or degrading treatment.

ND
Article 8 Right to Remedy

No discussion of legal remedies or right to effective remedy for violations.

ND
Article 9 No Arbitrary Detention

No discussion of arbitrary arrest or detention.

ND
Article 11 Presumption of Innocence

No discussion of presumption of innocence or burden of proof in criminal proceedings.

ND
Article 16 Marriage & Family

No discussion of rights to marry or found family.

ND
Article 30 No Destruction of Rights

No discussion of preventing destruction of UDHR or its rights through interpretation.

Structural Channel
What the site does
Element Modifier Affects Note
Legal & Terms
Privacy
No privacy policy or tracking disclosure observed on accessible pages.
Terms of Service
No terms of service accessible from navigation.
Identity & Mission
Mission +0.05
Article 19
Domain appears designed for independent technical commentary and free expression of ideas about software development. This supports editorial freedom without institutional constraint.
Editorial Code
No editorial code or standards statement observed.
Ownership
Individual-authored blog; clear authorship model supports accountability.
Access & Distribution
Access Model +0.10
Article 19 Article 26
Content appears freely accessible without paywall or registration, supporting open access to information and ideas.
Ad/Tracking
No advertising or tracking infrastructure visible on page.
Accessibility
No explicit accessibility statement observed.
br_tracking +0.05
Preamble ¶5 Article 12 Article 19
No third-party trackers detected
br_security -0.05
Article 3 Article 12
Security headers: HTTPS
br_accessibility -0.05
Article 26 Article 27 ¶1
No accessibility features detected
br_consent 0.00
Article 12 Article 19 Article 20 ¶2
No cookie consent banner detected
+0.20
Article 19 Freedom of Expression
High Advocacy Practice
Structural
+0.20
Context Modifier
+0.15
SETL
+0.23

Blog platform operates as free speech medium without apparent censorship or editorial gatekeeping. Content formatted to invite reader engagement and further research through linked references.

+0.15
Article 18 Freedom of Thought
High Advocacy Practice
Structural
+0.15
Context Modifier
0.00
SETL
+0.21

Blog platform provides direct outlet for dissenting technical opinion, with author maintaining independence from major AI companies and establishing freedom to publish critical analysis.

+0.15
Article 26 Education
High Advocacy Practice
Structural
+0.15
Context Modifier
+0.10
SETL
+0.21

Blog platform provides free access to technical education and critical analysis. Author freely shares personal experience, reasoning, and references to foundational texts enabling reader learning.

+0.10
Article 15 Nationality
Medium Framing Practice
Structural
+0.10
Context Modifier
0.00
SETL
+0.09

Blog addresses audience of software professionals, affirming their membership in technical community with shared standards and responsibilities.

+0.10
Article 22 Social Security
Medium Framing Practice
Structural
+0.10
Context Modifier
0.00
SETL
+0.19

Blog provides platform for technologists to discuss working conditions and sustainability of development practices.

+0.10
Article 27 Cultural Participation
High Advocacy Practice
Structural
+0.10
Context Modifier
0.00
SETL
+0.19

Blog participates in technical community discourse, with author engaging in and inviting community participation through references and gratitude to reviewers.

+0.10
Article 28 Social & International Order
Medium Framing Practice
Structural
+0.10
Context Modifier
0.00
SETL
+0.09

Blog operates as platform enabling discussion of software practices with implications for broader stakeholder ecosystem.

+0.05
Article 14 Asylum
Medium Framing Practice
Structural
+0.05
Context Modifier
0.00
SETL
+0.17

Blog platform provides space for refuge from mainstream LLM advocacy narratives, allowing expression of dissenting technical views without institutional pressure.

+0.05
Article 23 Work & Equal Pay
Medium Framing Practice
Structural
+0.05
Context Modifier
0.00
SETL
+0.17

Blog platform allows software professionals to articulate working conditions concerns and maintain voice in technological change affecting their labor.

ND
Preamble Preamble
Medium Framing

Content frames programming as intellectual work requiring human reasoning and collaboration, values dignity of human labor in software development, implicitly affirms human agency in work and decision-making.

ND
Article 1 Freedom, Equality, Brotherhood
Medium Framing

Content emphasizes equal reasoning capacity of humans and systems, argues against automation that undermines human judgment in software decisions. Advocates for human responsibility and deliberate choice in development practices.

ND
Article 2 Non-Discrimination
Low

Content does not directly address non-discrimination, though implicit in critique is concern that LLM productivity claims may be equally applied across different development contexts without acknowledging structural differences.

ND
Article 3 Life, Liberty, Security
Medium Framing

Content emphasizes right to meaningful work and security; critiques LLM productivity metrics that prioritize speed over correctness, stability, and sustainable practices. Argues that well-functioning software (security and reliability) requires human care.

ND
Article 4 No Slavery

No discussion of slavery or forced labor.

ND
Article 5 No Torture

No discussion of torture or degrading treatment.

ND
Article 6 Legal Personhood
Medium Framing

Content affirms right to recognition as a person; critiques LLM-centric narratives that diminish human programmer agency, reasoning, and recognition. Emphasizes that responsibility and credit remain with human practitioners.

ND
Article 7 Equality Before Law
Medium Framing

Content advocates for equal treatment in responsibility and code review. Argues that all developers, regardless of tool use, must maintain equal standards. Emphasizes peer review as fundamental practice.

ND
Article 8 Right to Remedy

No discussion of legal remedies or right to effective remedy for violations.

ND
Article 9 No Arbitrary Detention

No discussion of arbitrary arrest or detention.

ND
Article 10 Fair Hearing
Medium Framing

Content implicitly affirms fair hearing and due process in development practices; emphasizes transparent code review, explicit standards, and peer accountability. Critiques automation that bypasses human judgment in decisions affecting system behavior.

ND
Article 11 Presumption of Innocence

No discussion of presumption of innocence or burden of proof in criminal proceedings.

ND
Article 12 Privacy
Medium Framing

Content raises concerns about loss of privacy in rapid LLM-driven development; customers exposed to systems developed without full human understanding. Concerns about personal data handling when developers cannot fully explain system behavior.

ND
Article 13 Freedom of Movement
Low Framing

Content tangentially addresses freedom of movement and residence through critique of how LLM-driven development may create brittle, unmaintainable systems that constrain users' ability to migrate or modify software.

ND
Article 16 Marriage & Family

No discussion of rights to marry or found family.

ND
Article 17 Property
Low Framing

Content implicitly addresses property rights; argues against uncritical appropriation of LLM outputs and advocates for using existing libraries and established patterns rather than generating custom code, respecting established intellectual property.

ND
Article 20 Assembly & Association
Medium Framing

Content advocates for peaceful assembly and association in software development communities. Emphasizes community standards, peer review, and collective responsibility against isolation of individual productivity.

ND
Article 21 Political Participation
Low Framing

Content tangentially addresses participation in governance through emphasis on explicit choice and deliberate consideration of technological direction. Advocates against implicit adoption of practices without community deliberation.

ND
Article 24 Rest & Leisure
Medium Framing

Content advocates for rest, leisure, and reasonable working hours through critique of burnout-inducing acceleration. Emphasizes deliberate pacing of work and sustainability.

ND
Article 25 Standard of Living
Medium Framing

Content advocates for standard of living through emphasis on product quality, user welfare, and sustainable development. Critiques practices that compromise system reliability and user experience.

ND
Article 29 Duties to Community
Medium Framing

Content advocates for duties and responsibilities alongside rights; repeatedly emphasizes programmer accountability, peer responsibility, and community obligation.

ND
Article 30 No Destruction of Rights

No discussion of preventing destruction of UDHR or its rights through interpretation.

Supplementary Signals
How this content communicates, beyond directional lean. Learn more
Epistemic Quality
How well-sourced and evidence-based is this content?
0.77 medium claims
Sources
0.8
Evidence
0.8
Uncertainty
0.7
Purpose
0.8
Propaganda Flags
2 manipulative rhetoric techniques found
2 techniques detected
appeal to authority
Extensive use of quotes from Dijkstra, Ken Thompson, Bill Gates, Linus Torvalds to support claims about lines of code as poor metric. While acknowledged as 'appeals to authority' in appendix, used to support main arguments.
loaded language
Phrases like 'LLMs entice us with code too quickly. We are easily led,' 'false belief that lines of code mean anything,' characterizing certain practices with negative framing.
Emotional Tone
Emotional character: positive/negative, intensity, authority
measured
Valence
-0.3
Arousal
0.6
Dominance
0.6
Transparency
Does the content identify its author and disclose interests?
0.67
✓ Author ✓ Conflicts ✗ Funding
More signals: context, framing & audience
Solution Orientation
Does this content offer solutions or only describe problems?
0.59 mixed
Reader Agency
0.7
Stakeholder Voice
Whose perspectives are represented in this content?
0.55 4 perspectives
Speaks: individualsworkers
About: corporationinstitutionmarginalized
Temporal Framing
Is this content looking backward, at the present, or forward?
present medium term
Geographic Scope
What geographic area does this content cover?
global
Complexity
How accessible is this content to a general audience?
moderate medium jargon domain specific
Longitudinal 37 HN snapshots · 17 evals
+1 0 −1 HN
Audit Trail 37 entries
2026-03-16 00:31 eval_success PSQ evaluated: g-PSQ=0.323 (3 dims) - -
2026-03-16 00:31 eval Evaluated by llama-3.3-70b-wai-psq: +0.32 (Moderate positive)
2026-03-16 00:30 eval_success Lite evaluated: Neutral (-0.05) - -
2026-03-16 00:30 eval Evaluated by llama-3.3-70b-wai: -0.05 (Neutral)
reasoning
Technical post on codegen and productivity
2026-03-16 00:30 rater_validation_warn Lite validation warnings for model llama-3.3-70b-wai: 1W 0R - -
2026-03-15 23:42 eval_success PSQ evaluated: g-PSQ=0.280 (3 dims) - -
2026-03-15 23:42 eval Evaluated by llama-4-scout-wai-psq: +0.28 (Mild positive) 0.00
2026-03-15 23:10 eval_success Lite evaluated: Neutral (0.01) - -
2026-03-15 23:10 eval Evaluated by llama-4-scout-wai: +0.01 (Neutral) +0.09
reasoning
The content discusses the role of generative AI in software development, focusing on productivity and code generation. I
2026-03-15 23:10 rater_validation_warn Lite validation warnings for model llama-4-scout-wai: 0W 1R - -
2026-03-15 22:37 eval_success Evaluated: Mild positive (0.17) - -
2026-03-15 22:37 eval Evaluated by claude-haiku-4-5-20251001: +0.17 (Mild positive) 18,893 tokens
2026-03-15 22:37 rater_validation_warn Validation warnings for model claude-haiku-4-5-20251001: 7W 29R - -
2026-03-15 21:26 eval_success PSQ evaluated: g-PSQ=0.280 (3 dims) - -
2026-03-15 21:26 eval Evaluated by llama-4-scout-wai-psq: +0.28 (Mild positive) 0.00
2026-03-15 21:16 eval_success Lite evaluated: Neutral (-0.08) - -
2026-03-15 21:16 eval Evaluated by llama-4-scout-wai: -0.08 (Neutral) 0.00
reasoning
The content discusses the role of generative AI in software development, focusing on productivity and code generation. I
2026-03-15 21:16 rater_validation_warn Lite validation warnings for model llama-4-scout-wai: 1W 1R - -
2026-03-15 20:46 eval_success PSQ evaluated: g-PSQ=0.280 (3 dims) - -
2026-03-15 20:46 eval Evaluated by llama-4-scout-wai-psq: +0.28 (Mild positive) 0.00
2026-03-15 20:36 eval_success Lite evaluated: Neutral (-0.08) - -
2026-03-15 20:36 eval Evaluated by llama-4-scout-wai: -0.08 (Neutral) 0.00
reasoning
The content discusses the role of generative AI in software development, focusing on productivity and code generation. I
2026-03-15 20:36 rater_validation_warn Lite validation warnings for model llama-4-scout-wai: 1W 1R - -
2026-03-15 20:08 eval_success PSQ evaluated: g-PSQ=0.280 (3 dims) - -
2026-03-15 20:08 eval Evaluated by llama-4-scout-wai-psq: +0.28 (Mild positive) 0.00
2026-03-15 20:01 eval_success Lite evaluated: Neutral (-0.08) - -
2026-03-15 20:01 eval Evaluated by llama-4-scout-wai: -0.08 (Neutral) -0.06
reasoning
The content discusses the role of generative AI in software development, focusing on productivity and code generation. I
2026-03-15 20:01 rater_validation_warn Lite validation warnings for model llama-4-scout-wai: 1W 1R - -
2026-03-15 19:31 eval_success PSQ evaluated: g-PSQ=0.280 (3 dims) - -
2026-03-15 19:31 eval Evaluated by llama-4-scout-wai-psq: +0.28 (Mild positive) 0.00
2026-03-15 19:26 eval_success Lite evaluated: Neutral (-0.02) - -
2026-03-15 19:26 eval Evaluated by llama-4-scout-wai: -0.02 (Neutral) 0.00
reasoning
The content discusses the role of generative AI in software development, focusing on productivity and code generation. I
2026-03-15 19:26 rater_validation_warn Lite validation warnings for model llama-4-scout-wai: 0W 1R - -
2026-03-15 18:53 eval Evaluated by llama-4-scout-wai-psq: +0.28 (Mild positive) 0.00
2026-03-15 18:42 eval Evaluated by llama-4-scout-wai: -0.02 (Neutral) 0.00
reasoning
The content discusses the role of generative AI in software development, focusing on productivity and code generation. I
2026-03-15 17:48 eval Evaluated by llama-4-scout-wai-psq: +0.28 (Mild positive)
2026-03-15 17:32 eval Evaluated by llama-4-scout-wai: -0.02 (Neutral)
reasoning
The content discusses the role of generative AI in software development, focusing on productivity and code generation. I