69 stories tagged ai-coding-assistants, newest first.
3 points by KentBeck | 1 comments
Kent Beck newsletter post arguing AI coding "genies" should remove toil while preserving the productive struggle that builds skill, using a RoboGym analogy.
86 points by vaughands | 44 comments
Personal blog post by Vaughan Hilts describing how using LLMs reduced his typing and appears to have eased his long-standing repetitive strain injury.
11 points by aray07 | 8 comments
A personal blog post by a developer of 18 years explaining how Claude Code with Opus 4.5 led him to stop writing code by hand entirely.
73 points by haute_cuisine | 40 comments
A Substack essay by Alex Martsinovich arguing that agentic coding causes four problems: slop, alienation, deskilling, and team fallout.
148 points by saibotk | 153 comments
Personal blog post about a developer spending a month without AI tools at work, reflecting on how AI use made them lazier and a worse developer.
6 points by fscaramuzza | 6 comments
Communications of the ACM research article on the gap between individual efficiency gains and overall throughput when using GitHub Copilot; page showed a Cloudflare check.
![Screenshot of blog.szypowi.cz: Claude Code reads AGENTS.md only when telemetry is on [fixed]](https://hn.urlbox.com/shots/renders/2026/9/23/cca66142-3ad8-48cc-b5e2-08b9750af8e2.jpg)
453 points by pszypowicz | 257 comments
Blog post reporting that Claude Code 2.1.277's AGENTS.md support is gated behind a remote feature flag and skipped when telemetry is off, with measurements and a CLAUDE.md workaround.
6 points by meetpateltech | 0 comments
A blog post on jola.dev arguing that LLM assistants, by doing tasks for users instead of explaining them, undermine learning.
631 points by datadrivenangel | 224 comments
The official Claude Code changelog page listing release notes, new features, improvements, and bug fixes by version.
1072 points by matthieu_bl | 415 comments
Opusfived is a web game or interactive site challenging users to get Claude to make an "Add to Cart" button blue without changing anything else.
888 points by numeri | 799 comments
A blog post arguing that Anthropic's Opus 5 feels worse to work with than Opus 4.7, 4.8, and Fable because it makes assumptions and reinterprets plans without asking.
1195 points by cloudking | 501 comments
Ask HN thread asking whether anyone has fully replaced Claude or GPT with a locally run model for daily coding, and what hardware and token speeds they get.
480 points by prakashqwerty | 148 comments
A CLAUDE.md file in Stanford CS336's assignment repository giving guidelines for AI coding assistants working with students on the course.
340 points by andruby | 264 comments
A tweet thread by Aaron Perris showing Apple shipped CLAUDE.md files in Apple Support app v5.13, then pushed v5.13.1 to remove them.
1285 points by elmean | 696 comments
A tweet by Theo claiming Claude Code refuses requests or charges extra when commit messages mention "OpenClaw".
1189 points by homebrewer | 505 comments
GitHub issue reporting that the string HERMES.md in git commit history causes Claude Code to bill requests to extra usage instead of the Max plan quota.
697 points by frizlab | 509 comments
GitHub blog announcement that starting June 1, GitHub Copilot will move to usage-based billing, with usage consuming GitHub AI Credits.
336 points by speckx | 216 comments
Blog post arguing it's fine to use AI coding tools like Claude Code to finish abandoned side projects, describing building a YouTube Music to OpenSubsonic connector.
905 points by mfiguiere | 673 comments
Anthropic engineering post explaining how recent Claude Code quality complaints traced to three separate changes, and what the company is changing in response.
405 points by pella | 234 comments
Blog post describing the 'over-editing' problem, where AI coding models rewrite more code than needed when fixing simple bugs, with accompanying code.
499 points by zorrn | 204 comments
GitHub blog company-news post announcing changes to GitHub Copilot Individual plans, framed as ensuring a reliable, predictable experience for existing customers.
972 points by mikeevans | 517 comments
OpenAI product announcement of a major update to Codex, its AI coding agent, expanding its capabilities for its 3+ million developer users.
484 points by hmokiguess | 361 comments
Linux kernel documentation file giving guidance for AI tools and developers using AI assistance when contributing to the kernel.
1252 points by StanAngeloff | 686 comments
GitHub issue on anthropics/claude-code reporting that Claude Code became unusable for complex engineering tasks after the February model updates.
321 points by louiereederson | 199 comments
Claude's Code is a dashboard tracking GitHub commits made with Anthropic's Claude Code, showing growth rates, doubling time, and early adopter repositories.
1446 points by breton | 521 comments
A GitHub Gist showing a terminal screenshot where an AI coding assistant proposes a plan and asks "Shall I implement it?" and the user answers "No".
545 points by ndr42 | 435 comments
Ars Technica article (via Financial Times) reporting Amazon will require senior engineers to approve AI-assisted code changes after AWS outages linked to AI tools.
411 points by dnw | 368 comments
Substack essay from Vagabond Research arguing that LLMs produce plausible rather than correct code, using a simple database primary key lookup test as an example.
938 points by shannoncc | 802 comments
A Tell HN post by a 60-year-old developer saying Claude Code has rekindled the excitement he felt learning ASP, COM, and VB6, with reader comments.
343 points by saikatsg | 242 comments
Blog post by Ivan Turkovic arguing that AI tools make producing code easier while making the broader job of being a software engineer harder.
1060 points by cleak | 347 comments
Caleb Leak's blog post describing how he set up a system letting his dog 'vibe code' video games using AI coding tools.
994 points by WXLCKNO | 634 comments
Blog post on Symmetry Breaking arguing that Anthropic's Claude Code is being dumbed down, opening with a Simpsons "Am I out of touch?" meme.
372 points by fugu2 | 197 comments
Blog post by Tim Plaisted explaining how to connect Claude Code to a local open-source LLM when the usage quota is exhausted.
390 points by vismit2000 | 304 comments
Anthropic research post and paper examining how using AI assistance affects the development of coding skills.
472 points by AffableSpatula | 305 comments
An X post describing a hidden multi-agent 'Swarms' feature in Anthropic's Claude Code, linked alongside a GitHub tool for enabling unreleased features.
414 points by voxadam | 658 comments
IEEE Spectrum guest article by Jamie Twiss arguing newer AI coding assistants are more prone to silent but deadly failure modes.
503 points by KothuRoti | 346 comments
An X/Twitter post by Boris Cherny, creator of Claude Code, describing his personal Claude Code setup; the post content did not load.
376 points by martinald | 584 comments
Blog post by Martin Alderson arguing that agentic AI coding has dramatically cut the cost of building software and will transform the industry by 2026.
378 points by ayoisaiah | 130 comments
GitHub repository by Addy Osmani with a guide of about 30 pro tips for using Google's Gemini CLI for agentic coding.
440 points by sshh12 | 160 comments
A Substack post by Shrivu Shankar describing how he uses the various features of Anthropic's Claude Code for hobby and professional programming.
540 points by adocomplete | 343 comments
Anthropic product announcement dated Oct 20, 2025 introducing Claude Code on the web, a browser-based version of its AI coding tool.
453 points by todsacerdoti | 282 comments
A Codemanship blog post arguing that LLM-generated code creates 'comprehension debt' as teams accumulate code they don't understand.
618 points by chrisloy | 377 comments
Blog post by Chris Loy arguing that AI coding tools speed up typing code but risk neglecting the problem-solving that is the real work of software development.
370 points by elmsec | 395 comments
A personal blog post arguing that AI coding tools benefit senior developers more than juniors, outlining where AI helps and falls short in coding.
581 points by yoavfr | 238 comments
A humorous site that tracks and charts how many times per day Claude Code tells its user "You're absolutely right" or "right".
491 points by kmelve | 344 comments
Sanity blog post in which a staff engineer recounts six weeks of using Claude Code for development work, noting first attempts are mostly garbage.
463 points by zora_goron | 364 comments
Apple developer documentation page with the Xcode 26 Beta 7 release notes, covering new SDKs and features including Claude Sonnet integration.
378 points by floverfelt | 347 comments
Martin Fowler's blog post sharing scattered thoughts on the state of LLMs and AI and their effect on software development.
774 points by sixhobbits | 467 comments
A personal blog post describing how the author uses Claude Code for work, fun projects, and even as a text editor.
435 points by ingve | 180 comments
A blog post by Chris Dzombak sharing tips and practices for getting good results when using Anthropic's Claude Code AI coding tool.
864 points by coltonv | 601 comments
Blog post by Colton Voege arguing AI is not making engineers 10x more productive and offering advice for engineers feeling imposter syndrome over such claims.
543 points by mike1o1 | 543 comments
Puzzmo blog post by Orta Therox reflecting on six weeks of using Claude Code and how it changed his approach to writing and maintaining code.
345 points by mfiguiere | 213 comments
TechCrunch article reporting that Anthropic tightened usage limits for Claude Code without notifying users.
366 points by dejavucoder | 333 comments
A personal blog post recounting the author's experience using Anthropic's Claude Code coding assistant over two weeks.
322 points by sagacity | 194 comments
Zed blog post from its Agentic Engineering Sessions featuring engineer Alberto Fortin on why he scaled back his use of LLMs for software development.
332 points by ramoz | 140 comments
Anthropic documentation page explaining Claude Code hooks, user-defined shell commands that run at points in Claude Code's lifecycle.
400 points by vinhnx | 156 comments
Addy Osmani's Substack article offering a playbook of prompt engineering techniques for getting better results from AI coding assistants.
484 points by milkshakes | 687 comments
New York Times article reporting that Amazon software developers pushed to use AI say their work has become faster-paced and more like warehouse work.
640 points by swyx | 548 comments
Bloomberg news article reporting that OpenAI has agreed to acquire AI coding startup Windsurf for about $3 billion.
549 points by sqs | 232 comments
Anthropic engineering blog post sharing tips and best practices for using Claude Code, a command-line tool for agentic coding, across codebases and languages.
496 points by nomilk | 315 comments
A Cursor forum bug report where a user says the AI editor refused to generate more code past about 800 lines, telling them to learn coding themselves.
407 points by tosh | 451 comments
An X (Twitter) post by Steve Yegge sharing his impressions after using Anthropic's Claude Code for a couple of days; the post content did not load.
456 points by kiyanwang | 403 comments
An essay arguing that AI coding models stifle adoption of new technologies due to training data cutoffs and system prompt bias toward specific tools.
454 points by ahamez | 258 comments
Zed blog post announcing edit prediction in the Zed editor, powered by Zeta, a new open source model that anticipates the user's next code edit.
397 points by throwaway_43793 | 751 comments
An Ask HN thread in which a veteran software engineer asks how SWEs can future-proof their careers as LLMs increasingly write code.
436 points by 0x63_Problems | 213 comments
A Gauge blog article arguing that AI coding tools make technical debt more costly rather than less relevant.
380 points by tomyedwab | 338 comments
A blog post on Arguing with Algorithms reviewing the Cursor AI code editor and describing how the author uses it to write code.
418 points by BitWiseVibe | 275 comments
A Medium post by Tate Smith about frustration with fixing customers' AI-generated code, partially covered by a Medium sign-up modal.
418 points by alexmolas | 185 comments
Mistral AI blog post from May 29, 2024 announcing Codestral, its generative AI model for code.