Carsten Lawrenz

@gsd_foundation I plan to use it in my C10:OS for a non technical user to be able to one shot their application.

🚨 GSD 2.0 IS OUT TOMORROW AT 12:00PM CST 🚨 What's the first thing you're working on?

@gsd_foundation That’s all I need to se brother. I’ll plug that in when it’s ready

Soon

Claude code in Ghostty is the on-ramp for Agentic Engineering. C10:OS

@gsd_foundation I’m working on something that will help this effort. It’s coming together will have a beta soon. C10:OS

One of our top priorities at the GSD foundation is making it easier for agents to get set up and start using the GSD framework for reliable development. While GSD is currently optimised for human use, we will be shipping a version that is agent friendly soon. Exciting stuff 🪄

@official_taches Make it BOLD so that others can corporate in ther had built projects

I think it’s about time I release a manifesto for GSD. Above all, I will be honest and transparent. I will not rug you. I will not lie to you. I will not involve myself with people who do not have the projects best interests at heart. Full manifesto coming shortly - alongside the updated website. #makecryptowholesomeagain

@sized_in GSD is the real deal!

$GSD spent some time mapping who is actually adopting GSD and why. security engineers, enterprise ai, indie builders, designers, agent folks - all arriving at the same conclusion independently this isn’t hype-driven growth. it’s execution-driven adoption full write-up:

@official_taches Fish is faster and friendlier. Makes typing so easy. But you need to let go of old habits. Option-Key gives the most natural response. It also has a crazy built in history.

@Cars10 How have you set up your Ghostty? Also what is Fish Shell? As opposed to bash/zsh?

I live in the terminal these days. I love the terminal.

I've changed my whole MacBook setup removed all the junk. Different user ID's for different modes /Cars10 -> Vibe -> C10 ! No more noise, no more annoying ads. Three monitors now properly configured. so I don't need to wear my glasses. forget Stacks and Stage manager confusion. Fullscreen mode only when on laptop only incredible Smooth Scaling of text. The fonts are amazing!

We’re back baby

Adoption phase

Short answer: this is a very big deal — arguably the biggest step yet for $GSD. Let’s break it down plainly. When Lex Christopherson says: “GSD is getting a memory system” He’s not talking about simple chat history or logs. He means persistent, structured memory across builds. In practice, this likely means: - GSD will remember what worked in past sessions - It can reuse proven patterns, decisions, and workflows - Less re-explaining context every time you start a new task - Less “Claude amnesia” between sessions This is a direct attack on context rot. The “global hub” part (also important) “A global hub for monitoring and tracking your GSD sessions” That implies: - Visibility into past runs - Session analytics (what succeeded, what failed) - Possibly shared templates or internal benchmarks - A single control plane instead of scattered local runs Think: - GitHub Actions + Notion memory + agent telemetry…but specifically for how you build with GSD The real killer feature (this line matters most) “It’ll learn from what you did before that worked.” This is where things get spicy 🌶️ That suggests: - Pattern recognition across your builds - Reduced research time - Faster convergence on good solutions - GSD evolving with you, not just executing commands At that point, GSD stops being: “a smart meta-prompting system” And starts becoming: a personal build intelligence layer Why this is a BIG deal for GSD (contextually) Right now: - Claude Code = insanely powerful but forgetful - Most agent systems = stateless, noisy, brittle - “Multi-agent” hype = lots of motion, little learning With memory: - GSD compounds - Every project makes the next one easier - Skill accrues to you, not just the model - This is how tools become infrastructure, not toys This is exactly how: - IDEs beat text editors - Git beat zip files - CI beat “works on my machine” Strategic implication (zooming out) If done right, this puts GSD: - Ahead of vibe-coding - Ahead of agent swarms -,Ahead of prompt-of-the-week frameworks Because memory = leverage. Most tools reset every run.
GSD is signaling it wants to compound intelligence over time. That’s rare. And dangerous (in a good way). Bottom line How big is this? 🔹 Minor feature? ❌
🔹 Nice-to-have? ❌
🔹 Core differentiator? ✅
🔹 Long-term moat? ✅
🔹 “Oh shit” moment for competing tools? ✅ If GSD nails memory without turning into bloat, this is the moment it stops being “a framework” and starts being a system you don’t want to build without

@official_taches Well that takes a task off my list! Awesome to see this.

GSD now officially support Gemini CLI.

Announcing the launch of Hooks in Gemini CLI🚀🪝 Take full control and customize the agentic loop to your specific needs. Hooks allow you to add context, validate actions, enforce policies, loop the agent to continue to iterate, and a lot more! 👇Read the launch blog below

@arscontexta Some guy created this epic visual of the relationships between .md files made by my GSD system. It works in 3D so you can manipulate the thingamabob. Have you tried GSD yet btw?

TruePLM... the road is far but this thing should build itself! Stayed Tuned. GSD /Cars10

Open terminal and type: npx get-shit-done-cc Then launch Claude Code and type: /gsd:new-project See you in the future.

@karpathy Check out this silly experiment I did, a Sudoku Solver in 84 programming languages. I’ve experienced everything you mentioned, they are sneaky! linkedin.com/posts/carstenl…

A few random notes from claude coding quite a bit last few weeks. Coding workflow. Given the latest lift in LLM coding capability, like many others I rapidly went from about 80% manual+autocomplete coding and 20% agents in November to 80% agent coding and 20% edits+touchups in December. i.e. I really am mostly programming in English now, a bit sheepishly telling the LLM what code to write... in words. It hurts the ego a bit but the power to operate over software in large "code actions" is just too net useful, especially once you adapt to it, configure it, learn to use it, and wrap your head around what it can and cannot do. This is easily the biggest change to my basic coding workflow in ~2 decades of programming and it happened over the course of a few weeks. I'd expect something similar to be happening to well into double digit percent of engineers out there, while the awareness of it in the general population feels well into low single digit percent. IDEs/agent swarms/fallability. Both the "no need for IDE anymore" hype and the "agent swarm" hype is imo too much for right now. The models definitely still make mistakes and if you have any code you actually care about I would watch them like a hawk, in a nice large IDE on the side. The mistakes have changed a lot - they are not simple syntax errors anymore, they are subtle conceptual errors that a slightly sloppy, hasty junior dev might do. The most common category is that the models make wrong assumptions on your behalf and just run along with them without checking. They also don't manage their confusion, they don't seek clarifications, they don't surface inconsistencies, they don't present tradeoffs, they don't push back when they should, and they are still a little too sycophantic. Things get better in plan mode, but there is some need for a lightweight inline plan mode. They also really like to overcomplicate code and APIs, they bloat abstractions, they don't clean up dead code after themselves, etc. They will implement an inefficient, bloated, brittle construction over 1000 lines of code and it's up to you to be like "umm couldn't you just do this instead?" and they will be like "of course!" and immediately cut it down to 100 lines. They still sometimes change/remove comments and code they don't like or don't sufficiently understand as side effects, even if it is orthogonal to the task at hand. All of this happens despite a few simple attempts to fix it via instructions in CLAUDE . md. Despite all these issues, it is still a net huge improvement and it's very difficult to imagine going back to manual coding. TLDR everyone has their developing flow, my current is a small few CC sessions on the left in ghostty windows/tabs and an IDE on the right for viewing the code + manual edits. Tenacity. It's so interesting to watch an agent relentlessly work at something. They never get tired, they never get demoralized, they just keep going and trying things where a person would have given up long ago to fight another day. It's a "feel the AGI" moment to watch it struggle with something for a long time just to come out victorious 30 minutes later. You realize that stamina is a core bottleneck to work and that with LLMs in hand it has been dramatically increased. Speedups. It's not clear how to measure the "speedup" of LLM assistance. Certainly I feel net way faster at what I was going to do, but the main effect is that I do a lot more than I was going to do because 1) I can code up all kinds of things that just wouldn't have been worth coding before and 2) I can approach code that I couldn't work on before because of knowledge/skill issue. So certainly it's speedup, but it's possibly a lot more an expansion. Leverage. LLMs are exceptionally good at looping until they meet specific goals and this is where most of the "feel the AGI" magic is to be found. Don't tell it what to do, give it success criteria and watch it go. Get it to write tests first and then pass them. Put it in the loop with a browser MCP. Write the naive algorithm that is very likely correct first, then ask it to optimize it while preserving correctness. Change your approach from imperative to declarative to get the agents looping longer and gain leverage. Fun. I didn't anticipate that with agents programming feels *more* fun because a lot of the fill in the blanks drudgery is removed and what remains is the creative part. I also feel less blocked/stuck (which is not fun) and I experience a lot more courage because there's almost always a way to work hand in hand with it to make some positive progress. I have seen the opposite sentiment from other people too; LLM coding will split up engineers based on those who primarily liked coding and those who primarily liked building. Atrophy. I've already noticed that I am slowly starting to atrophy my ability to write code manually. Generation (writing code) and discrimination (reading code) are different capabilities in the brain. Largely due to all the little mostly syntactic details involved in programming, you can review code just fine even if you struggle to write it. Slopacolypse. I am bracing for 2026 as the year of the slopacolypse across all of github, substack, arxiv, X/instagram, and generally all digital media. We're also going to see a lot more AI hype productivity theater (is that even possible?), on the side of actual, real improvements. Questions. A few of the questions on my mind: - What happens to the "10X engineer" - the ratio of productivity between the mean and the max engineer? It's quite possible that this grows *a lot*. - Armed with LLMs, do generalists increasingly outperform specialists? LLMs are a lot better at fill in the blanks (the micro) than grand strategy (the macro). - What does LLM coding feel like in the future? Is it like playing StarCraft? Playing Factorio? Playing music? - How much of society is bottlenecked by digital knowledge work? TLDR Where does this leave us? LLM agent capabilities (Claude & Codex especially) have crossed some kind of threshold of coherence around December 2025 and caused a phase shift in software engineering and closely related. The intelligence part suddenly feels quite a bit ahead of all the rest of it - integrations (tools, knowledge), the necessity for new organizational workflows, processes, diffusion more generally. 2026 is going to be a high energy year as the industry metabolizes the new capability.

@finnbags @BagsApp That’s why we need to rally around independent developers who want to play with there thing full time like GSD!

Ralph demonstrated how powerful funding open source software can be on @BagsApp The coin reached a market cap of $50M and raised over $200K for the creator. However, despite understanding the opportunity and communicating regularly with our team, the creator of ralph decided to sell his tokens publicly after letting our team know that he would hold them. We did our best to offer solutions including a developer grant, vested supply, and an offer to buy his coins directly. The lesson we learned is that we need to protect traders even more by strengthening incentive alignment between developers and traders. We're going to double down on education and build systems to support developers who are committed to making their projects successful long term, rather than for a quick profit. To all the talented developers and traders coming into our ecosystem, we are here to support you. We are committed to constantly improving our product and we appreciate all your feedback. There's nothing that will stop us from funding and supporting the best builders in the world on @BagsApp bags.fm

@FFlyme31331 Every great Journey starts with the first step.....

@Cars10 I am encouraged, if you succeed, then this is a billion-dollar business, and most importantly, it is necessary for the market.

My post on LinkedIn about GSD. linkedin.com/posts/carstenl…