Ralph Wiggum Loop Goes Viral as Dario Predicts Full SWE Automation Within a Year

January 20, 2026 · 49 source posts

Daily Wrap-Up

Today felt like a tipping point in how developers talk about autonomous coding. The Ralph Wiggum loop, a pattern for running Claude Code in an unattended bash loop, went from niche power-user trick to mainstream discourse practically overnight. Seven separate posts in today's feed either praised it, taught it, or warned about its addictive qualities. When @mattpocockuk, one of the TypeScript community's most visible voices, says Ralph "makes Cursor advice feel quaint," you know the Overton window has shifted on what counts as normal developer workflow.

The other thread running through today's conversation was Dario Amodei's claim that AI models will handle "most, maybe all" of what software engineers do within 6 to 12 months. The reaction split predictably: some treated it as validation of the vibe coding movement, others pointed out that automating code generation doesn't mean automating software engineering. @GergelyOrosz offered the most grounded take, arguing that demand for engineers will increase, not decrease, because the bottleneck was never typing speed. The tension between "coding is dead" and "we need more engineers than ever" is going to define 2026's hiring conversations.

The most entertaining moment was easily @stevekrouse's extended rant comparing managing multiple Claude Code instances to 1950s TV producers putting radio announcers on camera. His point that humans can only hold seven things in working memory, and two concurrent agent sessions already max that out, is the kind of uncomfortable truth the "run 10 agents in parallel" crowd doesn't want to hear. The most practical takeaway for developers: if you're experimenting with autonomous coding loops, start by specifying your module structure and interfaces upfront before handing off to the agent. As @mattpocockuk learned the hard way, skipping that step produces slop, and the rework cost erases the speed gains.

Quick Hits

@vasuman shared a post titled "AI Agents 102," presumably a follow-up guide for those past the basics.
@parcadei dropped "WTF is a Context Graph?" calling it "a guide to the trillion-dollar problem."
@meowbooksj posted "top 10 IDE betrayals," which feels increasingly relatable in the age of AI-powered editors.
@excalidraw upgraded their text-to-diagram feature with a chat interface, streaming, and smarter generation.
@johnpalmer coined "Claude Cowork user" as a derogatory term, which is both funny and a sign of emerging tribal identity around tools.
@mikeishiring proposed that we'll soon have three identities: social networks, IRL, and an agent version of ourselves.
@shadcn highlighted a new way to turn code into shareable registry items, calling it something he'd been looking for.
@ryanflorence wondered why nothing on the mobile web is animated well, then said he's buying a course on it.
@bibryam shared lessons on writing a great CLAUDE.md file, drawn from analyzing over 2,500 repositories.
@denk_tweets celebrated beehiiv crossing $2M MRR and shared their early-days playbook of 10 tactics.
@EleanorKonik got Claude running inside Obsidian via the terminal plugin to help with creative writing research.
@framara launched TuCuento, an interactive storytelling app for parents and kids to create stories together.
@siavashg emerged from stealth with Stilla AI, a "Multiplayer AI" platform backed by $5M from General Catalyst.
@testingcatalog reported that X open-sourced their recommendation algorithm for transparency.
@steipete shared his PR review workflow, noting he's reviewed 1,000+ PRs and merged fewer than 10 without changes.
@aleenaamiir shared a detailed Gemini prompt for generating isometric 3D educational dioramas.
@weswinder fed the new X algorithm to Opus 4.5 and posted the recommended posting strategy for maximum reach.
@herkuch noted frustration with not being able to select models in OpenCode.
@steipete expressed continued amazement at @clawdbot making phone calls.
@ideabrowser argued that failed startup ideas from the graveyard could now succeed because solo builders have AI and don't need VC capital or teams.
@dom_lucre shared a viral post about an AI art demo that has digital artists worried about obsolescence before 2026 ends.

The Ralph Wiggum Loop Takes Over

The single biggest theme today was the Ralph Wiggum loop, an autonomous coding pattern where Claude Code runs in a bash loop, grinding through tasks while you sleep, eat, or do literally anything else. The conversation has moved well past "here's a cool trick" into territory that looks more like a movement.

@d4m1n captured the intensity: "I now run 1-2 loops 24/7, tweaking, iterating. Before sleep I set off a loop, I wake up 3-5x a night thinking of it with excitement." He called the pattern "extremely unhealthy" while simultaneously unable to stop, which is about as honest a product testimonial as you'll ever get. His other post was more declarative: "Using Ralph Wiggum loop will put you ahead of 98% of devs."

@Hesamation went further, comparing it to buying Bitcoin in 2012 and warning that "the window will be closed in just a few months." @paraddox described presenting the loop to two engineers still using VS Code AI extensions, where a 10x loop with GLM-4.7 fixed something their Opus setup couldn't. "They went quiet. Now I'm doing a workshop on it."

But @mattpocockuk offered the critical counterpoint that Ralph evangelists tend to skip: "Got a lot of slop out of Ralph today. The reason was, I didn't specify what modules I wanted up front, and that I wanted a simple, testable interface." He referenced "A Philosophy of Software Design," suggesting that the real skill isn't running the loop but knowing how to scope work for it. Meanwhile, his other post admitted that after discovering Ralph, conventional agent tips "feel a bit quaint" since "all this advice can be automated away with a few lines in a bash loop." @GeoffreyHuntley, who created the pattern, was in the replies directing people to his own resources.

The Great Agent Workflow Debate

Beyond Ralph specifically, today saw a rich conversation about how developers should actually work with coding agents. The opinions ranged from "run as many as possible" to "that's fundamentally wrong."

@stevekrouse delivered the day's most memorable rant against the multi-agent management approach: "Managing SUCKS. You're alienated from the work. Your feedback loops are terrible." He compared it to 1950s producers putting radio announcers on TV, missing that the medium demanded entirely new formats. His alternative vision: agents should passively ingest your GitHub repos, email, and commit history, then "ONLY NOTIFY ME WITH A FULLY WORKING PULL REQUEST, TOTALLY VERIFIED, THAT OTHER AGENTS HAVE REVIEWED."

On the practical end, @aye_aye_kaplan shared three concrete tips: always start in Plan Mode, start new chats frequently to avoid context pollution, and leverage AI for code review. @ericzakariasson added two more: "plan sync, implement async" and "create validation environments so you can ask the agent to check its own changes." @mntruell shared Cursor's official tips, which overlap heavily: plan first, use tests as feedback loops, revert and re-plan when things go sideways. @ctatedev provided living proof of the approach, claiming an entire weekend of agent-coding produced a complex networking and orchestration system that "would've taken 1-2 years solo." @techgirl1908 praised Cloudflare's Code Mode in Goose for cutting tokens and LLM calls in half.

Dario's 12-Month Prediction Sparks Debate

Anthropic CEO Dario Amodei predicted that AI will handle "most, maybe all" of what software engineers do end-to-end within 6 to 12 months. Both @WesRoth and @slow_developer amplified the quote, with @slow_developer adding important context: "We're approaching a feedback loop where AI builds better AI. But the loop isn't fully closed yet, chip manufacturing and training time still limit speed."

@GergelyOrosz pushed back on the doom interpretation: "That us engineers will not write most (or any) code by hand doesn't mean what many replies assume it does, that there won't be demand for SWEs. The opposite: I expect more demand for software engineers who can build reliable+complex software with LLMs!" This tracks with everything else in today's feed. The developers thriving with Ralph and agent loops aren't less skilled than traditional coders. They're applying engineering judgment at a higher level of abstraction.

@Hesamation and @thekitze both referenced the Node.js creator's recent comments about the end of hand-written code, with @Hesamation noting that "you can be emotional about coding by hand and insist that AI coding sucks, but it doesn't make you any less delusional." Harsh, but the signal from today's posts is clear: the debate isn't whether agents will write code, it's how fast developers can learn to direct them effectively.

Skills Are the New MCP

Remotion made a splash today by launching Agent Skills for video creation, letting developers make videos entirely through Claude Code prompts. @Remotion announced it with a demo video created purely by prompting, and @andrewqu confirmed the quality: "I nearly 1-shotted this launch video." @JNYBGR echoed the sentiment: "Created this video without writing any code, but also without needing After Effects skills. Yet I was able to control every detail!"

The bigger story might be @intellectronica's declaration that they've dropped all MCP servers from their local setup, replacing Context7, Tavily, and Playwright with Skills and curl. "SKILLs are all you need!" This suggests the developer tooling ecosystem is consolidating around Skills as the preferred integration pattern, at least for local development workflows. Skills are lighter weight, don't require running background servers, and compose more naturally with agent loops. If this trend holds, MCP's role may narrow to enterprise and cloud-hosted scenarios where the server model makes more sense.

Claude Code Ecosystem Expands

Anthropic shipped two notable updates today. The VS Code extension for Claude Code hit general availability, bringing the CLI experience into the editor with @-mention file context, slash commands, and other familiar features. @claudeai also announced four health data integrations in beta: Apple Health, Health Connect, HealthEx, and Function Health. The health integrations signal Anthropic's push beyond developer tools into consumer-facing capabilities.

On the GitHub side, @GHchangelog announced that GitHub Copilot now supports OpenCode's open-source agent with no additional license needed. The walls between coding agent ecosystems continue to come down.

The AI Adoption Gap

While developer Twitter debated autonomous coding loops, @bwarrn dropped a reality check from lunch with a Fortune 500 AI consultant: "Some of the biggest companies on earth use zero AI tools. Not even ChatGPT. Execs only recognize: ChatGPT, Copilot, Gemini. Everyone feels behind. Nobody knows what to buy." The "AI saturation" narrative, he argued, is a Silicon Valley bubble. "Rest of the world hasn't started yet."

Palantir CEO Alex Karp, as quoted by @jawwwn_, reinforced this from a different angle: "If you just buy LLMs off the shelf and try to do any of this, it won't work. It's not precise enough." His argument that you need "a software layer to orchestrate and manage the LLMs in a language your enterprise understands" to create real value is essentially the enterprise version of what the Ralph Wiggum community is doing with bash scripts and CLAUDE.md files. The tools differ wildly, but the principle is the same: raw model access isn't enough. You need orchestration.

Source Posts

Dom Lucre | Breaker of Narratives @dom_lucre · Jan 20

🔥🚨BREAKING: Digital artists are in a panic after this creator showed the current power of creating art with the help of AI which has digital artists fearing that they could become obsolete before 2026 is over. https://t.co/OQPucL74SR

Idea Browser @ideabrowser · Jan 20

I bet this graveyard has 100+ ideas that would make $3M per year. Just because they failed, doesn’t mean you would. They needed VC capital They needed a team. They needed to be in Silicon Valley. They needed $1B valuations You don’t need that. You have AI now. You have better tools. Good luck I’m rotting for you.

a aditya @adxtyahq

Someone curated 925 failed VC-backed startups, broke down why they failed, and how to make it work with today’s tech - https://t.co/NFUhrhe7P2 Cool fr🙌 https://t.co/vOv2fUDnhY

geoff @GeoffreyHuntley · Jan 20

@damianplayer skip this and learn from me (i created ralph) https://t.co/zDb4V4xw8s

ℏ

ℏεsam @Hesamation · Jan 20

when the creator of node.js says the era of humans writing code is over, just one week after Linus tries out vibe coding, you know a chapter in technology is slowly closing to give way for a new one. you can be emotional about coding by hand and insist that AI coding sucks, but it doesn’t make you any less delusional.

R Ryan Dahl @rough__sea

This has been said a thousand times before, but allow me to add my own voice: the era of humans writing code is over. Disturbing for those of us who identify as SWEs, but no less true. That's not to say SWEs don't have work to do, but writing syntax directly is not it.

Dan ⚡️ @d4m1n · Jan 20

Using Ralph Wiggum loop will put you ahead of 98% of devs

This is for devs. It's for those that like to create, to break down problems and cannot help themselves from coming up with ideas hour by hour. You ke...

Jon Kaplan @aye_aye_kaplan · Jan 20

My top 3 tips for coding with agents: 1. Always start with Plan Mode. It's better to iterate in natural language and then execute once you know what the agent is going to do. This will save you time, effort, and tokens! 2. Start new chats frequently. Remember that your role is to point the Agent in the right direction to make the changes you need. If you change topics, the context window will get muddied. You will also be spending more tokens on longer chats. 3. Leverage AI to do your code review. If you know the failure case, ask a model. One prompt I often use is "scan the changes on my branch and confirm nothing is impacted outside of my feature flag". As a safety net for everything outside this issues-you-expect umbrella, use Bugbot.

dei @parcadei · Jan 20

WTF is a Context Graph? A Guide to the Trillion-Dollar Problem

You’ve read 15 articles, skimmed 12 threads and watched 4 podcasts but you still can't answer "what is a context graph?" Allow me to confuse you furth...

Remotion @Remotion · Jan 20

Here's how we created the above video! Full prompt history: https://t.co/OhyuqqsD0o https://t.co/h1T4JwCIKS

eric zakariasson @ericzakariasson · Jan 20

exactly how i code with agents. some more: 4. plan sync, implement async. if you can quickly align on a plan, you'll have higher confidence when handing off to a cloud agent 5. create validation environments, so you can ask agent to check its own changes

J Jon Kaplan @aye_aye_kaplan

Steve Krouse @stevekrouse · Jan 20

MANAGING LOTS OF CLAUDE CODES IS SUPER DUMB That's like in the 1950s thinking that TV is just radio announcers at a desk reading from a script. Nope. It's sitcoms, movies, YouTube, TikTok. Or in the 1970s thinking that the future of accounting would be managing a bunch of number crunching "agents". Nope. It's Excel or Quickbooks. Managing SUCKS. You're alienated from the work. Your feedback loops are terrible. What's better? Being a craftsperson with a powerful tool My brother in christ, you can only think of 7 things at a time, and if you're running 2 Claude Codes, each has a couple details that need your attention, so you're already all maxed out of things to think about, so you can't even notice how un-productive you're being Yes, I get the instinct to RUN AS MUCH INFERENCE AS POSSIBLE LLMs seem like super cheap employees. If you aren't giving them the MAXIMUM work, you're leaving money on the table I have a suggestion for you. A way for you to run LOTS OF INFERENCE. Let's go back to my boy @worrydream INTERACTIVITY CONSIDERED HARMFUL There is so much context stored in my github repo, my issues, my commit history, also my email inbox, etc. If you could somehow be passively ingesting all that and running all sort of inference on it WITHOUT ME HAVING TO MANAGE IT, that sounds awesome There is so much cleaning up that I'd love someone to do on my GitHub Issues backlog Or if you want to go ahead and try to end-to-end solve some of my tickets and ONLY NOTIFY ME WITH A FULLY WORKING PULL REQUEST, TOTALLY VERIFIED, THAT OTHER AGENTS HAVE REVIEWED, WITH AN AMAZING PR EXPLAINER THATS SUPER CONCISE AND NOT SLOP, oh my god, take my money Can you do something similar for my email inbox? I'll name my first born after you. I want LESS management. LESS slop. If I wanted more management and more slop, I would hire interns or offshore contractors I hire the best engineers I can find who give me less to manage, less to edit their writing I want AI to do the same

Angie Jones @techgirl1908 · Jan 20

Cloudflare was definitely on to something with Code Mode. I tried it in goose this weekend for very extensive work. Half the tokens, messages, and LLM calls! This means lower costs, longer sessions, and less back and forth with the agent. I'm sold! https://t.co/gUEUPxmYU9

Peter Steinberger 🦞 @steipete · Jan 20

Still amazed every time @clawdbot does a phone call.

S Swatantra Sohni @TheGeneralistHQ

All thanks to @steipete https://t.co/GQ3ZJNF1Tj

ℏ

ℏεsam @Hesamation · Jan 20

learning Ralph might be equivalent of buying Bitcoin in 2012. only this time the window will be closed in just a few months. this is a great article if you’re wondering wtf Ralph is.

D Damian Player @damianplayer

the people learning this now will be untouchable in 3 months.

Andrew Qu @andrewqu · Jan 20

Wow this skill by remotion is SICK I nearly 1 shotted this launch video for https://t.co/kdOWe32i5V @opencode chat transcript down below 👇 https://t.co/eNJkC6HoZt

R Remotion @Remotion

Remotion now has Agent Skills - make videos just with Claude Code! $ npx skills add remotion-dev/skills This animation was created just by prompting 👇 https://t.co/hadnkHlG6E

kitze 🚀 @thekitze · Jan 20

node js creator: coding is dead avg mid miderson: i will never trust llms!!! I need hours of convincing to help my skill issues

R Ryan Dahl @rough__sea

vas @vasuman · Jan 20

AI Agents 102

So, you've built an agent that kinda barely works in demos and now you want to deploy it for real users with real use cases. Demos and production have...

Matt Pocock @mattpocockuk · Jan 20

Got a lot of slop out of Ralph today The reason was, I didn't specify what modules I wanted up front, and that I wanted a simple, testable interface From the Philosophy of Software Design: https://t.co/jJcGHyH9Nd

Jonny Burger @JNYBGR · Jan 20

Created this video without writing any code, but also without needing After Effects skills. Yet I was able to control every detail! Never felt so powerful 🧙🏻

R Remotion @Remotion

Remotion now has Agent Skills - make videos just with Claude Code! $ npx skills add remotion-dev/skills This animation was created just by prompting 👇 https://t.co/hadnkHlG6E

Ryan Florence @ryanflorence · Jan 20

why is nothing on the mobile web animated like this? anyway, i'm going to buy this course

E Emil Kowalski @emilkowalski

You can enroll in my animation course for the next 10 days! It's the perfect way to learn the theory behind great animations, but also how to build them in code. Now with a skill file for agents. We'll cover all of these components and more, source code included. https://t.co/RvK4piO5QQ

Claude @claudeai · Jan 20

Claude can now securely connect to your health data. Four new integrations are now available in beta: Apple Health (iOS), Health Connect (Android), HealthEx, and Function Health. https://t.co/tTCnxOGt7i

Excalidraw @excalidraw · Jan 20

We've made text-to-diagram better. Chat interface. Streaming. Smarter. Faster. Stronger. https://t.co/q0xKW3dJTi

GitHub Changelog @GHchangelog · Jan 20

GitHub Copilot now supports OpenCode's open source agent. No additional license needed. https://t.co/yfMJnw1Gg5

Wes Winder @weswinder · Jan 20

i asked opus 4.5 to analyze the new x algorithm this is the posting strategy it recommended follow this for maximum reach https://t.co/QuXFCk92Sx

E Engineering @XEng

We have open-sourced our new 𝕏 algorithm, powered by the same transformer architecture as xAI's Grok model. Check it out here: https://t.co/3WKwZkdgmB

Jawwwn @jawwwn_ · Jan 20

Palantir CEO Alex Karp says people think we’re in an AI bubble because a lot of AI just doesn’t work: “If you just buy LLMs off the shelf and try to do any of this, it won’t work.” “It’s not precise enough. You can’t do underwriting. You can’t do these things that are regulated.” “People have tried things that just can never work. You buy a LLM, put it on your stack, and wonder why it’s not working.” “What you’re going to see, especially in America, is people trying to do something like Ontology by hand.” “Once you build a software layer to orchestrate and manage the LLMs in a language your enterprise understands, you actually can create value.” “There’s a lot of discussion on if we’re in an AI bubble. What is the meaning of this bubble? If anything, we’re just in a lag. There’s a lot of AI, some of it works.” “Go back to the battlefield context: everybody in the world assumed this would not work. But now it does work. Now the question is, ‘How can I get it to work for my country?’” “Palantir barely has a sales force. In fact, it seems to be getting smaller and smaller every time I go see them.”

Eleanor Konik @EleanorKonik · Jan 20

I gotta say, I am surprised at how easy the terminal plugin was to install for Obsidian. Now I've got Claude cooking on finding the stuff I remember that's related to the (very) short story I wrote last night, & am ready to put together my list of things to do for the day. https://t.co/Da2rtb2sF4

Bilgin Ibryam @bibryam · Jan 20

How to write a great https://t.co/iJwx0WHbIw: Lessons from over 2,500 repositories https://t.co/Y0eOtkYw9W

Claude @claudeai · Jan 20

The VS Code extension for Claude Code is now generally available. It’s now much closer to the CLI experience: @-mention files for context, use familiar slash commands (/model, /mcp, /context), and more. Download it here: https://t.co/q95Cw4soMk https://t.co/3BCWPvybdZ

Ben @bwarrn · Jan 20

Lunch w/ an exited founder who helps fortune 500 companies adopt AI. Insane reality check: Some of the biggest companies on earth use *zero* AI tools. Not even ChatGPT. Execs only recognize: ChatGPT, Copilot, Gemini (maybe Perplexity). Everyone feels behind. Nobody knows what to buy or how to plug it in. The "AI saturation" narrative is another example of what a bubble Silicon Valley is. Rest of the world hasn’t started yet. We have to build for the 99%.

Matt Pocock @mattpocockuk · Jan 20

It's crazy how after discovering Ralph this stuff feels a bit quaint All this advice can be automated away with a few lines in a bash loop

J Jon Kaplan @aye_aye_kaplan

Tyler Denk 🐝 @denk_tweets · Jan 20

beehiiv just crossed $2M MRR but most founders are still stuck trying to generate their first $100K to celebrate the milestone I’m going to try something new… I’m sharing the exact playbook we used in the early days (10 simple tactics) hope this helps someone: https://t.co/Cwaos2LXgB

Peter Steinberger 🦞 @steipete · Jan 20

How I start pretty much every PR review. (Yeah, I could do a slash command but speaking is so fast and I usually already have thoughts that it doesn't make me faster) Of the 1000+ PRs I reviewed so far I merged <10 without changes, often massively so. https://t.co/SzFWAHgIIh

Haider. @slow_developer · Jan 20

Anthropic CEO, Dario Amodei: "we might be 6-12 months away from models doing all of what software engineers do end-to-end" We're approaching a feedback loop where AI builds better AI But the loop isn't fully closed yet, chip manufacturing and training time still limit speed

shadcn @shadcn · Jan 20

Finally, a way to quickly turn code into shareable registry items. Been looking for this.

R Ronny Badilla @rbadillap

@shadcn you asked for it, you got it. 🚀 Announcing pastecn. A simple way to store your snippets and instantly get a shadcn-compatible registry URL. No setup. Just paste and ship. ⚡ https://t.co/mT3Ydr0DAy

Ddox @paraddox · Jan 20

Presented the Ralph loop to 2 engineers still using VS Code AI extensions. My 10x loops GLM-4.7 fixed something their Opus didn't. They went quiet. Now I'm doing a workshop on it. No good deed goes unpunished.

Remotion @Remotion · Jan 20

Remotion now has Agent Skills - make videos just with Claude Code! $ npx skills add remotion-dev/skills This animation was created just by prompting 👇 https://t.co/hadnkHlG6E

Gergely Orosz @GergelyOrosz · Jan 20

That us engineers will not write most (or any) code by hand doesn’t mean what many replies assume it does - that there won’t be demand for SWEs. The opposite: I expect more demand for software engineers who can build reliable+complex software with LLMs! https://t.co/sSfyCm8jk8 https://t.co/0JSgdHxlXr

R Ryan Dahl @rough__sea

TestingCatalog News 🗞 @testingcatalog · Jan 20

BREAKING 🚨: The X algorithm has been open-sourced by the X team to let users observe how it evolves transparently. https://t.co/96IKQCs58m

E Engineering @XEng

We have open-sourced our new 𝕏 algorithm, powered by the same transformer architecture as xAI's Grok model. Check it out here: https://t.co/3WKwZkdgmB

Aleena Amir @aleenaamiir · Jan 20

“How It Works” Educational Dioramas Gemini Nano Banana Pro Prompt: Create a clear, 45° top-down isometric miniature 3D educational diorama explaining [PROCESS / CONCEPT]. Use soft refined textures, realistic PBR materials, and gentle lifelike lighting. Build a stepped or layered diorama base showing each stage of the process with subtle arrows or paths. Include tiny stylized figures interacting with each stage (no facial details). Use a clean solid [BACKGROUND COLOR] background. At the top-center, display [PROCESS NAME] in large bold text, directly beneath it show a short explanation subtitle, and place a minimal symbolic icon below. All text must automatically match the background contrast (white or black).

John Palmer @johnpalmer · Jan 20

you kinda seem more like a Claude Cowork user (derogatory)

Dan ⚡️ @d4m1n · Jan 20

Ralph loop is extremely unhealthy. I now run 1-2 loops 24/7, tweaking, iterating. Before sleep I set off a loop, I wake up 3-5x a night thinking of it with excitement. My workflow is getting better, but it's a steep learning curve. I'll share it all when I'm a bit further, there are some things I still don't like. The best thing? I am doing projects I never had time to do and I can't stop. I am tired too 😅 But having something code while you sleep is pretty incredible.

D Dan ⚡️ @d4m1n

Using Ralph Wiggum loop will put you ahead of 98% of devs

Michael Truell @mntruell · Jan 20

Our tips on how to use Cursor: - Start with a plan (Shift+Tab Plan Mode) - Let Cursor search on its own, don't over-tag context - Use tests as the feedback loop (TDD + iterate until green) - When it goes sideways: revert → tighten the plan → rerun - Keep long chats short; use @ Past Chats for continuity - Add lightweight .cursor/rules for recurring mistakes - Use skills + hooks for long-running "grind until tests pass" loops - Run multiple agents/models in parallel via worktrees

C Cursor @cursor_ai

Here's what we've learned from building and using coding agents. https://t.co/PuBtYuhyhd

Eleanor Berger @intellectronica · Jan 20

I am no longer using _any_ MCP servers in my local setup [ @code / @GitHubCopilot, @opencode, @claudeai code ]. ・Context7 → SKILL + curl ・Tavily → SKILL + curl ・Playwright → SKILL + agent-browser SKILLs are all you need!

Paco @framara · Jan 20

Today I’m launching TuCuento. An interactive storytelling app designed for parents and children to create stories together. Choose characters, make decisions as a team, and shape how the adventure unfolds. Demo video below 👇 Link in replies. https://t.co/u3EeI9q1Dz

Mikeishiring ⚡️🤖 @mikeishiring · Jan 20

@marckohlbrugge Seconded on the YOLO but instead you should create a new layer for it to interact with. E.g an email for the bot which you forward things onto. This will probably be the future, we'll have 3 identities: - Social networks - IRL -Agent version of you

Khushal ☘️ @herkuch · Jan 20

I tried this but couldn't find option to choose model with opencode :(

A Addy Osmani @addyosmani

Vibe Kanban: orchestrate multiple AI coding agents in parallel. Free and 100% open-source. Switch between Claude Code, Codex Gemini CLI, and track task status from a single dashboard. https://t.co/XfZLWpevqM

Wes Roth @WesRoth · Jan 20

"Software Engineering Will Be Automatable in 12 Months," Anthropic CEO Dario Amodei predicts that AI models will be able to do 'most, maybe all' of what software engineers do end-to-end within 6 to 12 months, shifting engineers to editors. https://t.co/7bI7JmTtsb

Meowbooks @meowbooksj · Jan 20

top 10 IDE betrayals https://t.co/bKcEl2ziMk

C Claude @claudeai

Siavash @siavashg · Jan 20

AI made everyone 10x faster. But the faster individuals move, the harder it is to move together. Speed ≠ Progress when no one has the full picture. Today we emerge from stealth with $5M led by @GeneralCatalyst to fix this. Meet @stillaai : The first Multiplayer AI. 🧵 https://t.co/lkRsBjhCIt

Chris Tate @ctatedev · Jan 20

This is where we're at rn: I spent the 3-day weekend agent-coding a complex system: advanced networking, orchestration, caching, bare metal, reverse proxies, custom Linux kernel This would've taken me 1-2 years solo And the result might be one of the best in its category