+0.31 An AI agent coding skeptic tries AI agent coding, in excessive detail (minimaxir.com S:+0.21 )
56 points by minimaxir 2 days ago | 9 comments on HN | Moderate positive Contested Editorial · v3.7 · 2026-02-28 11:57:06 0
Summary Free Expression & Knowledge Access Advocates
A technical blog post documenting one developer's methodical evaluation of LLM agent capabilities, emphasizing skepticism toward automation hype and human critical evaluation. The content advocates for transparency in AI development, open knowledge sharing via open-source code, and preservation of human agency in technological augmentation—embodying values of free expression, scientific advancement access, and worker dignity.
Article Heatmap
Preamble: +0.16 — Preamble P Article 1: +0.12 — Freedom, Equality, Brotherhood 1 Article 2: ND — Non-Discrimination Article 2: No Data — Non-Discrimination 2 Article 3: ND — Life, Liberty, Security Article 3: No Data — 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: ND — Legal Personhood Article 6: No Data — Legal Personhood 6 Article 7: ND — Equality Before Law Article 7: No Data — 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: ND — Fair Hearing Article 10: No Data — Fair Hearing 10 Article 11: ND — Presumption of Innocence Article 11: No Data — Presumption of Innocence 11 Article 12: ND — Privacy Article 12: No Data — Privacy 12 Article 13: ND — Freedom of Movement Article 13: No Data — Freedom of Movement 13 Article 14: ND — Asylum Article 14: No Data — Asylum 14 Article 15: ND — Nationality Article 15: No Data — Nationality 15 Article 16: ND — Marriage & Family Article 16: No Data — Marriage & Family 16 Article 17: ND — Property Article 17: No Data — Property 17 Article 18: ND — Freedom of Thought Article 18: No Data — Freedom of Thought 18 Article 19: +0.46 — Freedom of Expression 19 Article 20: ND — Assembly & Association Article 20: No Data — Assembly & Association 20 Article 21: ND — Political Participation Article 21: No Data — Political Participation 21 Article 22: ND — Social Security Article 22: No Data — Social Security 22 Article 23: +0.11 — Work & Equal Pay 23 Article 24: ND — Rest & Leisure Article 24: No Data — Rest & Leisure 24 Article 25: ND — Standard of Living Article 25: No Data — Standard of Living 25 Article 26: ND — Education Article 26: No Data — Education 26 Article 27: +0.50 — Cultural Participation 27 Article 28: ND — Social & International Order Article 28: No Data — Social & International Order 28 Article 29: ND — Duties to Community Article 29: No Data — 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
Editorial Mean +0.31 Structural Mean +0.21
Weighted Mean +0.32 Unweighted Mean +0.27
Max +0.50 Article 27 Min +0.11 Article 23
Signal 5 No Data 26
Volatility 0.17 (Medium)
Negative 0 Channels E: 0.6 S: 0.4
SETL ℹ +0.14 Editorial-dominant
FW Ratio ℹ 55% 12 facts · 10 inferences
Evidence 11% coverage ℹ
2H 2M 1L 26 ND
Theme Radar
Foundation Security Legal Privacy & Movement Personal Expression Economic & Social Cultural Order & Duties Foundation: 0.14 (2 articles) Security: 0.00 (0 articles) Legal: 0.00 (0 articles) Privacy & Movement: 0.00 (0 articles) Personal: 0.00 (0 articles) Expression: 0.46 (1 articles) Economic & Social: 0.11 (1 articles) Cultural: 0.50 (1 articles) Order & Duties: 0.00 (0 articles)
HN Discussion 6 top-level · 1 replies
simonw 2026-02-27 20:51 UTC link
This is my favorite yet of the genre of "OK, coding agents got good in November" posts. It starts with relatively simple examples (YouTube metadata scraping) and by the end Max is rewriting Python's skikit-learn framework in Rust and making it way faster.
7777777phil 2026-02-27 21:18 UTC link
The sharp results all came from pairing domain expertise with detailed AGENTS.md files. The impressive Rust output happened because someone who knows Rust was steering it. Vague prompts got mediocre output. A model on its own converges to the mean of its training data, which is why the "vibe code everything" thesis keeps not holding up: https://philippdubach.com/posts/the-impossible-backhand/
busssard 2026-02-27 21:56 UTC link
this post reflects my experience with the model...
ej88 2026-02-27 21:58 UTC link
Thanks Max! This was a really interesting article and closely matches my own experience with how the agents have been progressing

one of the takeaways I get when reading skilled engineers' experiences with these tools is that they essentially offer leverage, and the more skill someone already has the higher their ceiling will be

verdverm 2026-02-27 22:12 UTC link
I second that spending effort on your AGENTS.md is game changing. Don't auto generate these, work with them and learn how to make them good (sparknotes and table of contents, keep minimal, distribute over dirs)
ivraatiems 2026-02-27 23:46 UTC link
This investigation aligns with my experience in a lot of ways. I'm con-the influence and behavior big AI companies, but lukewarm-to-pro the actual use of the technology itself.

I use Claude and other models frequently (mostly via Cursor, with a smattering of other tools) in my work now. It is not at the "I never write code myself" point, but the AI tools are absolutely capable of generating highly effective and usable code, usually nearly as good or as good as what I'd do myself, with guidance.

It hasn't eliminated the need for my existence as an engineer, but it has changed it drastically. It is much more like "tell the computer what I want and mostly get it" than it was a year ago.

And yet, I have friends and colleagues who reject it out of hand as useless, and are so skeptical of it that they suggest it must only be good because my skills are poor, or our codebase is bad, or I'm getting lucky.

I just can't totally credit any of those explanations anymore.

rudiksz 2026-02-28 09:19 UTC link
Claude isn't generating code that is "highly effective and usable code". I'm not your friend but I also reject your claims out of hand because I've also seen what Claude can and can't do.
Editorial Channel
What the content says
+0.50
Article 19 Freedom of Expression
High A: free expression and thought P: open knowledge dissemination
Editorial
+0.50
SETL
+0.22

Content is direct exercise of freedom of expression and thought. Author freely publishes opinion, analysis, and critique of dominant narratives without self-censorship. Shares detailed technical knowledge and code, facilitating others' access to information and ability to participate in scientific advancement.

+0.50
Article 27 Cultural Participation
High A: scientific advancement and knowledge access P: open authorship and intellectual property respect
Editorial
+0.50
SETL
0.00

Content directly promotes participation in scientific and technical advancement. Author shares detailed methodology, code, and results openly. Respects authorship by clearly attributing work and citing sources. Facilitates others' participation in technological development through open-source distribution.

+0.20
Preamble Preamble
Medium A: human dignity in human-AI collaboration F: skepticism toward technological determinism
Editorial
+0.20
SETL
+0.14

Content affirms human dignity and agency by critiquing hype narratives that position agents as replacing human judgment. Emphasizes human oversight ('manually reviewed'), critical evaluation, and sovereignty of human decision-making ('impugn the sovereignty of the human soul').

+0.20
Article 1 Freedom, Equality, Brotherhood
Medium A: human reason and critical thinking F: defense against technological determinism
Editorial
+0.20
SETL
+0.20

Content defends human reason and critical evaluation against passive acceptance of technological narratives. Author positions skepticism as intellectual integrity ('I actually don't use Generative LLMs often'), resisting social pressure to adopt tools uncritically.

+0.15
Article 23 Work & Equal Pay
Low F: worker dignity in human-agent collaboration
Editorial
+0.15
SETL
+0.12

Implicit engagement: Content discusses 'productivity gain' and optimization of human work through agents, framed within context of human oversight and skill maintenance. Author's concern that agents might lead to 'atrophy of programming skills' reflects attention to workers' right to develop abilities and maintain meaningful work.

ND
Article 2 Non-Discrimination

No observable content regarding freedom from discrimination.

ND
Article 3 Life, Liberty, Security

No observable content regarding right to life, liberty, personal security.

ND
Article 4 No Slavery

No observable content regarding slavery.

ND
Article 5 No Torture

No observable content regarding torture or cruel treatment.

ND
Article 6 Legal Personhood

No observable content regarding right to recognition as person before law.

ND
Article 7 Equality Before Law

No observable content regarding equality before law.

ND
Article 8 Right to Remedy

No observable content regarding effective remedy.

ND
Article 9 No Arbitrary Detention

No observable content regarding arrest and detention.

ND
Article 10 Fair Hearing

No observable content regarding right to fair trial.

ND
Article 11 Presumption of Innocence

No observable content regarding presumption of innocence.

ND
Article 12 Privacy

No observable content regarding privacy.

ND
Article 13 Freedom of Movement

No observable content regarding freedom of movement.

ND
Article 14 Asylum

No observable content regarding asylum.

ND
Article 15 Nationality

No observable content regarding nationality.

ND
Article 16 Marriage & Family

No observable content regarding family and marriage.

ND
Article 17 Property

No observable content regarding property.

ND
Article 18 Freedom of Thought

No observable content regarding freedom of thought, conscience, religion.

ND
Article 20 Assembly & Association

No observable content regarding assembly or association.

ND
Article 21 Political Participation

No observable content regarding political participation.

ND
Article 22 Social Security

No observable content regarding social security.

ND
Article 24 Rest & Leisure

No observable content regarding rest and leisure.

ND
Article 25 Standard of Living

No observable content regarding standard of living or health.

ND
Article 26 Education

No observable content regarding education.

ND
Article 28 Social & International Order

No observable content regarding social/international order.

ND
Article 29 Duties to Community

No observable content regarding duties to community.

ND
Article 30 No Destruction of Rights

No observable content regarding limitation of rights.
Structural Channel
What the site does
+0.50
Article 27 Cultural Participation
High A: scientific advancement and knowledge access P: open authorship and intellectual property respect
Structural
+0.50
Context Modifier
ND
SETL
0.00

GitHub repositories provide open-source code under permissive licenses. Attribution clear throughout (author name, publication date, citations). Multiple links enable direct access to underlying work and allow reuse, modification, and redistribution by others.

+0.40
Article 19 Freedom of Expression
High A: free expression and thought P: open knowledge dissemination
Structural
+0.40
Context Modifier
ND
SETL
+0.22

Blog platform provides unmediated publishing capability. GitHub links enable free access to open-source code. No paywalls, gating, or censorship signals. Content is discoverable and shareable.

+0.10
Preamble Preamble
Medium A: human dignity in human-AI collaboration F: skepticism toward technological determinism
Structural
+0.10
Context Modifier
ND
SETL
+0.14

Open blog platform facilitates unmediated expression. No barriers to author's communication.

+0.05
Article 23 Work & Equal Pay
Low F: worker dignity in human-agent collaboration
Structural
+0.05
Context Modifier
ND
SETL
+0.12

No direct structural signals regarding working conditions.

0.00
Article 1 Freedom, Equality, Brotherhood
Medium A: human reason and critical thinking F: defense against technological determinism
Structural
0.00
Context Modifier
ND
SETL
+0.20

No structural signals specific to this article.

ND
Article 2 Non-Discrimination

No structural signals.

ND
Article 3 Life, Liberty, Security

No structural signals.

ND
Article 4 No Slavery

No structural signals.

ND
Article 5 No Torture

No structural signals.

ND
Article 6 Legal Personhood

No structural signals.

ND
Article 7 Equality Before Law

No structural signals.

ND
Article 8 Right to Remedy

No structural signals.

ND
Article 9 No Arbitrary Detention

No structural signals.

ND
Article 10 Fair Hearing

No structural signals.

ND
Article 11 Presumption of Innocence

No structural signals.

ND
Article 12 Privacy

No structural signals.

ND
Article 13 Freedom of Movement

No structural signals.

ND
Article 14 Asylum

No structural signals.

ND
Article 15 Nationality

No structural signals.

ND
Article 16 Marriage & Family

No structural signals.

ND
Article 17 Property

No structural signals.

ND
Article 18 Freedom of Thought

No structural signals.

ND
Article 20 Assembly & Association

No structural signals.

ND
Article 21 Political Participation

No structural signals.

ND
Article 22 Social Security

No structural signals.

ND
Article 24 Rest & Leisure

No structural signals.

ND
Article 25 Standard of Living

No structural signals.

ND
Article 26 Education

No structural signals.

ND
Article 28 Social & International Order

No structural signals.

ND
Article 29 Duties to Community

No structural signals.

ND
Article 30 No Destruction of Rights

No structural signals.
Supplementary Signals
How this content communicates, beyond directional lean. Learn more
Epistemic Quality ℹ
How well-sourced and evidence-based is this content?
0.77 medium claims
Sources
0.8
Evidence
0.8
Uncertainty
0.7
Purpose
0.9
Propaganda Flags ℹ
No manipulative rhetoric detected
0 techniques detected
Emotional Tone ℹ
Emotional character: positive/negative, intensity, authority
measured
Valence
+0.5
Arousal
0.3
Dominance
0.6
Transparency ℹ
Does the content identify its author and disclose interests?
0.50
✓ Author
More signals: context, framing & audience
Solution Orientation ℹ
Does this content offer solutions or only describe problems?
0.85 solution oriented
Reader Agency
0.8
Stakeholder Voice ℹ
Whose perspectives are represented in this content?
0.42 3 perspectives
Speaks: individualscorporation
About: workersinstitution
Temporal Framing ℹ
Is this content looking backward, at the present, or forward?
mixed long term
Geographic Scope ℹ
What geographic area does this content cover?
global
San Francisco, United States
Complexity ℹ
How accessible is this content to a general audience?
technical high jargon domain specific
Longitudinal 75 HN snapshots · 54 evals
+1 0 −1 HN
Audit Trail 74 entries
2026-03-02 08:17 eval_success Evaluated: Neutral (0.00) - -
2026-03-02 08:17 model_divergence Cross-model spread 0.32 exceeds threshold (3 models) - -
2026-03-02 08:17 eval Evaluated by deepseek-v3.2: +0.00 (Neutral) 17,891 tokens -0.30
2026-03-01 18:48 eval_success Evaluated: Moderate positive (0.30) - -
2026-03-01 18:48 model_divergence Cross-model spread 0.32 exceeds threshold (3 models) - -
2026-03-01 18:48 eval Evaluated by deepseek-v3.2: +0.30 (Moderate positive) 12,717 tokens +0.20
2026-03-01 18:48 rater_validation_warn Validation warnings for model deepseek-v3.2: 0W 2R - -
2026-03-01 17:37 eval_success Evaluated: Neutral (0.10) - -
2026-03-01 17:37 model_divergence Cross-model spread 0.32 exceeds threshold (3 models) - -
2026-03-01 17:37 eval Evaluated by deepseek-v3.2: +0.10 (Neutral) 13,498 tokens -0.16
2026-02-28 17:11 eval_success Lite evaluated: Neutral (0.00) - -
2026-02-28 17:11 model_divergence Cross-model spread 0.32 exceeds threshold (3 models) - -
2026-02-28 17:11 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 15:38 model_divergence Cross-model spread 0.32 exceeds threshold (3 models) - -
2026-02-28 15:38 eval_success Lite evaluated: Neutral (0.00) - -
2026-02-28 15:38 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 15:26 model_divergence Cross-model spread 0.32 exceeds threshold (3 models) - -
2026-02-28 15:26 eval_success Lite evaluated: Neutral (0.00) - -
2026-02-28 15:26 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 13:57 eval_success Lite evaluated: Neutral (0.00) - -
2026-02-28 13:57 model_divergence Cross-model spread 0.32 exceeds threshold (3 models) - -
2026-02-28 13:57 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 13:26 model_divergence Cross-model spread 0.32 exceeds threshold (3 models) - -
2026-02-28 13:26 eval_success Lite evaluated: Neutral (0.00) - -
2026-02-28 13:26 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 13:07 eval_success Lite evaluated: Neutral (0.00) - -
2026-02-28 13:07 rater_validation_warn Lite validation warnings for model llama-3.3-70b-wai: 0W 1R - -
2026-02-28 13:07 model_divergence Cross-model spread 0.32 exceeds threshold (3 models) - -
2026-02-28 13:07 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 13:02 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 11:57 eval Evaluated by claude-haiku-4-5-20251001: +0.32 (Moderate positive)
2026-02-28 11:27 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 10:56 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 10:38 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 09:14 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 08:51 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 08:46 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 08:19 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 08:17 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 08:14 eval Evaluated by deepseek-v3.2: +0.26 (Mild positive) 13,551 tokens +0.21
2026-02-28 07:30 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 07:02 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 07:00 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 06:43 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 06:42 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 06:29 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 06:11 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 05:48 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 05:41 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 05:26 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 05:24 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 04:30 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 04:18 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 04:17 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 03:48 eval Evaluated by deepseek-v3.2: +0.05 (Neutral) 13,708 tokens
2026-02-28 03:41 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 03:36 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 03:14 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 03:12 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 03:02 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 02:40 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 02:38 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 02:36 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 02:36 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 02:13 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 01:58 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 01:54 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 01:51 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 01:38 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 01:30 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 01:30 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral) 0.00
reasoning
ED, neutral AI tech exploration
2026-02-28 01:29 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral) 0.00
reasoning
Technical blog post
2026-02-28 01:19 eval Evaluated by llama-3.3-70b-wai: 0.00 (Neutral)
reasoning
Technical blog post
2026-02-28 00:44 eval Evaluated by llama-4-scout-wai: 0.00 (Neutral)
reasoning
ED, neutral AI tech exploration