#ClaudeCode

Nick RadcliffeTop tip for working with Claude Code if you find that its auto-compactification is worse than a full frontal lobotomy (as I do).I have created an alias for starting it that sets a 30-min timer. When it goes off, I usually check context and if necessary capture state, restart, and import captured state.A lot of process, but way better than autocompactification for me.<a href="https://mathstodon.xyz/tags/llm" class="mention hashtag" rel="nofollow noopener" target="_blank">#llm</a> <a href="https://mathstodon.xyz/tags/claudecode" class="mention hashtag" rel="nofollow noopener" target="_blank">#claudecode</a> <a href="https://mathstodon.xyz/tags/CHOP" class="mention hashtag" rel="nofollow noopener" target="_blank">#CHOP</a>

Miguel Afonso Caetano"For a while now I’ve been hearing from engineers who run multiple coding agents at once—firing up several Claude Code or Codex CLI instances at the same time, sometimes in the same repo, sometimes against multiple checkouts or git worktrees.I was pretty skeptical about this at first. AI-generated code needs to be reviewed, which means the natural bottleneck on all of this is how fast I can review the results. It’s tough keeping up with just a single LLM given how fast they can churn things out, where’s the benefit from running more than one at a time if it just leaves me further behind?Despite my misgivings, over the past few weeks I’ve noticed myself quietly starting to embrace the parallel coding agent lifestyle.I can only focus on reviewing and landing one significant change at a time, but I’m finding an increasing number of tasks that can still be fired off in parallel without adding too much cognitive overhead to my primary work.Here are some patterns I’ve found for applying parallel agents effectively."<a href="https://simonwillison.net/2025/Oct/5/parallel-coding-agents/" rel="nofollow noopener" translate="no" target="_blank">https://simonwillison.net/2025/Oct/5/parallel-coding-agents/</a><a href="https://tldr.nettime.org/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI</a> <a href="https://tldr.nettime.org/tags/GenerativeAI" class="mention hashtag" rel="nofollow noopener" target="_blank">#GenerativeAI</a> <a href="https://tldr.nettime.org/tags/AIAgents" class="mention hashtag" rel="nofollow noopener" target="_blank">#AIAgents</a> <a href="https://tldr.nettime.org/tags/AgenticAI" class="mention hashtag" rel="nofollow noopener" target="_blank">#AgenticAI</a> <a href="https://tldr.nettime.org/tags/LLMs" class="mention hashtag" rel="nofollow noopener" target="_blank">#LLMs</a> <a href="https://tldr.nettime.org/tags/ClaudeCode" class="mention hashtag" rel="nofollow noopener" target="_blank">#ClaudeCode</a> <a href="https://tldr.nettime.org/tags/CodexCLI" class="mention hashtag" rel="nofollow noopener" target="_blank">#CodexCLI</a>

Dusk To Don :raccoon:<a href="https://metasocial.com/@jesse" class="u-url mention" rel="nofollow noopener" target="_blank">@jesse</a> Found your blogpost via simonwillison[.]net and it’s great—thanks for sharing!<a href="https://blog.fsck.com/2025/10/05/how-im-using-coding-agents-in-september-2025/?utm_source=substack&utm_medium=email" rel="nofollow noopener" translate="no" target="_blank">https://blog.fsck.com/2025/10/05/how-im-using-coding-agents-in-september-2025/?utm_source=substack&utm_medium=email</a><a href="https://todon.eu/tags/claudeCode" class="mention hashtag" rel="nofollow noopener" target="_blank">#claudeCode</a>

AI SparkupLLM 생성 코드의 숨은 함정 ‘이해 부채’ – 스펙 중심 개발로 해결하기 AI가 생성한 코드를 이해하지 못해 발생하는 '이해 부채' 문제와 이를 해결하는 스펙 중심 개발 방식을 실제 사례와 함께 소개합니다. GitHub과 개발자들이 제안하는 실용적인 워크플로우를 배워보세요. <a href="https://aisparkup.com/posts/5284" rel="nofollow noopener" translate="no" target="_blank">https://aisparkup.com/posts/5284</a>

Ma’mounYOU ARE ABSOLUTELY RIGHT!!!!! <a href="https://mastodon.social/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#LLM</a> <a href="https://mastodon.social/tags/ClaudeCode" class="mention hashtag" rel="nofollow noopener" target="_blank">#ClaudeCode</a>

Jeff TriplettThis Tier ranking is really interesting to me. For me:S-tier: Claude Code A-tier: Codex (arguably S-tier) B-tier, but on the verge of A-tier is Copilot CI for me. I haven't used Opencode or Cursor's CI product to rank yet, but this video has my attention. <a href="https://mastodon.social/tags/ClaudeCode" class="mention hashtag" rel="nofollow noopener" target="_blank">#ClaudeCode</a> <a href="https://mastodon.social/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#LLM</a> <a href="https://mastodon.social/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI</a> <a href="https://www.youtube.com/watch?v=tCGju2JB5Fw" rel="nofollow noopener" translate="no" target="_blank">https://www.youtube.com/watch?v=tCGju2JB5Fw</a>

Lorenzo 'kelset' SciandraI've been using heavily <a href="https://mastodon.online/tags/ClaudeCode" class="mention hashtag" rel="nofollow noopener" target="_blank">#ClaudeCode</a> for coding over the last month or so, I think I found my groove and approach to make it work consistently decently - here's a quick&dirty graph that shows my current flow:(ofc feedback super welcome 🙏)

Lorenzo 'kelset' Sciandrakind of wild that <a href="https://mastodon.online/tags/ClaudeCode" class="mention hashtag" rel="nofollow noopener" target="_blank">#ClaudeCode</a> is made with TS + React + Yoga 🤯🔗 <a href="https://newsletter.pragmaticengineer.com/p/how-claude-code-is-built" rel="nofollow noopener" translate="no" target="_blank">https://newsletter.pragmaticengineer.com/p/how-claude-code-is-built</a>

Konstantin 🔭Yeah, me too <a href="https://hachyderm.io/tags/ClaudeCode" class="mention hashtag" rel="nofollow noopener" target="_blank">#ClaudeCode</a>

Lorenzo 'kelset' SciandraThe more I play with <a href="https://mastodon.online/tags/ClaudeCode" class="mention hashtag" rel="nofollow noopener" target="_blank">#ClaudeCode</a>'s subagents, the more my ⚙️ spin - I think even more than ever being great at context engineering and apply that to subagents will really unlock the next "phase" of ai codingZach Wills has a great post for inspo: <a href="https://zachwills.net/how-to-use-claude-code-subagents-to-parallelize-development/" rel="nofollow noopener" translate="no" target="_blank">https://zachwills.net/how-to-use-claude-code-subagents-to-parallelize-development/</a>

Artem R 🇺🇦The fast that AI can get silently dumber over time is mind boggling. You choose some system at one point in time based on your needs and tests, and then slowly degrades <a href="https://www.anthropic.com/engineering/a-postmortem-of-three-recent-issues" rel="nofollow noopener" translate="no" target="_blank">https://www.anthropic.com/engineering/a-postmortem-of-three-recent-issues</a> <a href="https://indieweb.social/tags/ArtificialIntelligence" class="mention hashtag" rel="nofollow noopener" target="_blank">#ArtificialIntelligence</a> <a href="https://indieweb.social/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI</a> <a href="https://indieweb.social/tags/claude" class="mention hashtag" rel="nofollow noopener" target="_blank">#claude</a> <a href="https://indieweb.social/tags/claudecode" class="mention hashtag" rel="nofollow noopener" target="_blank">#claudecode</a>

Ma’mounAI was completely stuck and going in circles trying to help me align the labels of SwiftUI buttons in an HStack. I pointed it to <a href="https://www.swiftuifieldguide.com" rel="nofollow noopener" translate="no" target="_blank">https://www.swiftuifieldguide.com</a> and told it to use the knowledge and wisdom to align the views, and voila... it came up with the perfect (and pretty sophisticated yet simple) solution! 🤯What frustrated me is that I still don't understand SwiftUI even after seeing the code! 😂<a href="https://mastodon.social/tags/claudecode" class="mention hashtag" rel="nofollow noopener" target="_blank">#claudecode</a> <a href="https://mastodon.social/tags/SwiftUI" class="mention hashtag" rel="nofollow noopener" target="_blank">#SwiftUI</a>

Python for Data ScienceWe have now described how to create a configuration for Claude Code so that it uses uv reliably: <a href="https://python4data.science/en/latest/productive/envs/uv/claude.html" rel="nofollow noopener" translate="no" target="_blank">https://python4data.science/en/latest/productive/envs/uv/claude.html</a> <a href="https://threads.net/@claudeai/" class="u-url mention" rel="nofollow noopener" target="_blank">@claudeai</a> <a href="https://mastodon.social/tags/ClaudeCode" class="mention hashtag" rel="nofollow noopener" target="_blank">#ClaudeCode</a> <a href="https://mastodon.social/tags/Python" class="mention hashtag" rel="nofollow noopener" target="_blank">#Python</a> <a href="https://mastodon.social/tags/Packaging" class="mention hashtag" rel="nofollow noopener" target="_blank">#Packaging</a> <a href="https://mastodon.social/tags/uv" class="mention hashtag" rel="nofollow noopener" target="_blank">#uv</a>

alexanderadamI can't believe that this is what we came to. Did any of the sci-fi authors anticipate what we arrived to?<a href="https://ruby.social/tags/claude_code" class="mention hashtag" rel="nofollow noopener" target="_blank">#claude_code</a> <a href="https://ruby.social/tags/claude4" class="mention hashtag" rel="nofollow noopener" target="_blank">#claude4</a> <a href="https://ruby.social/tags/llm" class="mention hashtag" rel="nofollow noopener" target="_blank">#llm</a> <a href="https://ruby.social/tags/vibe_coding" class="mention hashtag" rel="nofollow noopener" target="_blank">#vibe_coding</a> <a href="https://ruby.social/tags/VibeCoding" class="mention hashtag" rel="nofollow noopener" target="_blank">#VibeCoding</a> <a href="https://ruby.social/tags/ClaudeCode" class="mention hashtag" rel="nofollow noopener" target="_blank">#ClaudeCode</a> <a href="https://ruby.social/tags/ChatGPT" class="mention hashtag" rel="nofollow noopener" target="_blank">#ChatGPT</a> <a href="https://ruby.social/tags/ClaudeAI" class="mention hashtag" rel="nofollow noopener" target="_blank">#ClaudeAI</a> <a href="https://ruby.social/tags/programming" class="mention hashtag" rel="nofollow noopener" target="_blank">#programming</a> <a href="https://ruby.social/tags/ai" class="mention hashtag" rel="nofollow noopener" target="_blank">#ai</a>

WinbuzzerAnthropic Admits Three Infrastructure Bugs Caused Claude’s Performance Issues, Denies Throttling<a href="https://mastodon.social/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI</a> <a href="https://mastodon.social/tags/Anthropic" class="mention hashtag" rel="nofollow noopener" target="_blank">#Anthropic</a> <a href="https://mastodon.social/tags/Claude" class="mention hashtag" rel="nofollow noopener" target="_blank">#Claude</a> <a href="https://mastodon.social/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#LLM</a> <a href="https://mastodon.social/tags/CLaudeCode" class="mention hashtag" rel="nofollow noopener" target="_blank">#CLaudeCode</a><a href="https://winbuzzer.com/2025/09/18/anthropic-admits-three-infrastructure-bugs-caused-claudes-performance-issues-denies-throttling-xcxwbn" rel="nofollow noopener" translate="no" target="_blank">https://winbuzzer.com/2025/09/18/anthropic-admits-three-infrastructure-bugs-caused-claudes-performance-issues-denies-throttling-xcxwbn</a>

Jeff C. 🇺🇦Naughty language involving a butt

Jeff C. 🇺🇦"If you submit the merge request, you're responsible for the code change as if you typed it yourself" is the way to go with these sorts of things.Use AI for code generation? Fine. I use it, too. But I'm for damned sure going to look over the output and pay attention to what it's doing.If I can't explain what it's doing I'm not merging it.And that goes triple for any code that some other poor sap is going to have to reason through.<a href="https://indieweb.social/tags/LLM" class="mention hashtag" rel="nofollow noopener" target="_blank">#LLM</a> <a href="https://indieweb.social/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI</a> <a href="https://indieweb.social/tags/ClaudeCode" class="mention hashtag" rel="nofollow noopener" target="_blank">#ClaudeCode</a> <a href="https://social.treehouse.systems/@gfxstrand/115220843956925235" rel="nofollow noopener" translate="no" target="_blank">https://social.treehouse.systems/@gfxstrand/115220843956925235</a>

PeterBFinally I made some progress on the sorting machine overhaul. Sadly Claude was kinda lazy so you'll have to wait a bit longer for the details. Read the full update in my latest newsletter <a href="https://ml.streamhead.com/preview/445139/emails/165093726828364862" rel="nofollow noopener" translate="no" target="_blank">https://ml.streamhead.com/preview/445139/emails/165093726828364862</a>Sign up for the actual code and building plans right here <a href="https://subscribepage.io/sorting-updates" rel="nofollow noopener" translate="no" target="_blank">https://subscribepage.io/sorting-updates</a><a href="https://indiehackers.social/tags/lego" class="mention hashtag" rel="nofollow noopener" target="_blank">#lego</a> <a href="https://indiehackers.social/tags/poweredup" class="mention hashtag" rel="nofollow noopener" target="_blank">#poweredup</a> <a href="https://indiehackers.social/tags/ai" class="mention hashtag" rel="nofollow noopener" target="_blank">#ai</a> <a href="https://indiehackers.social/tags/python" class="mention hashtag" rel="nofollow noopener" target="_blank">#python</a> <a href="https://indiehackers.social/tags/claudecode" class="mention hashtag" rel="nofollow noopener" target="_blank">#claudecode</a>

Marius (windsheep) :donor:The switch from <a href="https://infosec.exchange/tags/claudecode" class="mention hashtag" rel="nofollow noopener" target="_blank">#claudecode</a> to Open-Source tools and models works fine for me.1. The CLI client I use is llxprt . This is a fork of the <a href="https://infosec.exchange/tags/gemini" class="mention hashtag" rel="nofollow noopener" target="_blank">#gemini</a> cli . <a href="https://github.com/acoliver/llxprt-code" rel="nofollow noopener" translate="no" target="_blank">https://github.com/acoliver/llxprt-code</a>The Desktop client is <a href="https://infosec.exchange/tags/anythingllm" class="mention hashtag" rel="nofollow noopener" target="_blank">#anythingllm</a> <a href="https://anythingllm.com/" rel="nofollow noopener" translate="no" target="_blank">https://anythingllm.com/</a>Both have MCP support, and work for macOS and Linux.2. For large code bases I am using ck search in my prompts and make llxpert to use it via Bash tool calls.<a href="https://github.com/BeaconBay/ck" rel="nofollow noopener" translate="no" target="_blank">https://github.com/BeaconBay/ck</a> My experiences with claude-context are that it's expensive and not effective for the cost (embeddings):<a href="https://github.com/zilliztech/claude-context" rel="nofollow noopener" translate="no" target="_blank">https://github.com/zilliztech/claude-context</a> I stopped using it. 3. I mostly use <a href="https://infosec.exchange/tags/OpenRouter" class="mention hashtag" rel="nofollow noopener" target="_blank">#OpenRouter</a> gpt-oss 120b as a coding modelkimi-k2-0905 or glm-4.5 for select tasks, like finding a function or simple test output summaries.These models are on par with <a href="https://infosec.exchange/tags/sonnet" class="mention hashtag" rel="nofollow noopener" target="_blank">#sonnet</a> or <a href="https://infosec.exchange/tags/opus" class="mention hashtag" rel="nofollow noopener" target="_blank">#opus</a> <a href="https://artificialanalysis.ai/models/glm-4.5?model-filters=open-source" rel="nofollow noopener" translate="no" target="_blank">https://artificialanalysis.ai/models/glm-4.5?model-filters=open-source</a>4. I mostly optimize my prompts, tools, and workflows. I write my own MCP tools, and use specific RAG approaches to establish the context.5. I self-host more and more. But I haven't found my stack here yet. I actually want gpt-oss, but it's too expensive. 6. <a href="https://infosec.exchange/tags/openai" class="mention hashtag" rel="nofollow noopener" target="_blank">#openai</a> codex cli is a myth to me. <a href="https://github.com/openai/codex" rel="nofollow noopener" translate="no" target="_blank">https://github.com/openai/codex</a>Yes, it's slow. But in numerous instances the output is great. It can be used with the same OpenRouter models.

Miguel Afonso Caetano"The Model Context Protocol (MCP) can empower LLM agents with potentially hundreds of tools to solve real-world tasks. But how do we make those tools maximally effective?In this post, we describe our most effective techniques for improving performance in a variety of agentic AI systems1.We begin by covering how you can:- Build and test prototypes of your tools - Create and run comprehensive evaluations of your tools with agents - Collaborate with agents like Claude Code to automatically increase the performance of your toolsWe conclude with key principles for writing high-quality tools we’ve identified along the way:- Choosing the right tools to implement (and not to implement) - Namespacing tools to define clear boundaries in functionality - Returning meaningful context from tools back to agents - Optimizing tool responses for token efficiency - Prompt-engineering tool descriptions and specs"<a href="https://www.anthropic.com/engineering/writing-tools-for-agents" rel="nofollow noopener" translate="no" target="_blank">https://www.anthropic.com/engineering/writing-tools-for-agents</a><a href="https://tldr.nettime.org/tags/AI" class="mention hashtag" rel="nofollow noopener" target="_blank">#AI</a> <a href="https://tldr.nettime.org/tags/GenerativeAI" class="mention hashtag" rel="nofollow noopener" target="_blank">#GenerativeAI</a> <a href="https://tldr.nettime.org/tags/AgenticAI" class="mention hashtag" rel="nofollow noopener" target="_blank">#AgenticAI</a> <a href="https://tldr.nettime.org/tags/MCP" class="mention hashtag" rel="nofollow noopener" target="_blank">#MCP</a> <a href="https://tldr.nettime.org/tags/LLMs" class="mention hashtag" rel="nofollow noopener" target="_blank">#LLMs</a> <a href="https://tldr.nettime.org/tags/Chatbots" class="mention hashtag" rel="nofollow noopener" target="_blank">#Chatbots</a> <a href="https://tldr.nettime.org/tags/PromptEngineering" class="mention hashtag" rel="nofollow noopener" target="_blank">#PromptEngineering</a> <a href="https://tldr.nettime.org/tags/Anthropic" class="mention hashtag" rel="nofollow noopener" target="_blank">#Anthropic</a> <a href="https://tldr.nettime.org/tags/Claude" class="mention hashtag" rel="nofollow noopener" target="_blank">#Claude</a> <a href="https://tldr.nettime.org/tags/ClaudeCode" class="mention hashtag" rel="nofollow noopener" target="_blank">#ClaudeCode</a>

Recent searches

Search options

Administered by:

Server stats: