Ron Kurti

150 posts

Ron Kurti banner
Ron Kurti

Ron Kurti

@RonKurti

I make technology friendly. Head of Product https://t.co/9RtSBCtv2C + https://t.co/r5ebvjDkPI Building https://t.co/sxVaAJMyCH on the side.

Los Angeles, CA เข้าร่วม Nisan 2008
270 กำลังติดตาม520 ผู้ติดตาม
fuho
fuho@fuho·
@RonKurti This is really good, thanks for creating it and letting people use it.
English
0
0
1
21
Ron Kurti
Ron Kurti@RonKurti·
Started repo-explainer.com 2 weeks ago. Yesterday I had to add category filters and lazy loading, because there are just too many repos to list. What next?
Ron Kurti tweet mediaRon Kurti tweet media
English
5
0
6
102
Bindu Reddy
Bindu Reddy@bindureddy·
LLMs are slowly automating white-collar work, even as haters continue to hate 🤷‍♀️ In 2026, they will - - create new features, test them, and release to prod - run vibe-marketing campaigns and run social media accounts - analyze data, generate reports, and come up with insights - run support, legal, procurement, and operations in general Literally everyone will feel the AGI 😍
English
67
20
293
15.6K
Shubham Saboo
Shubham Saboo@Saboo_Shubham_·
Self-healing Agent Skills. This is the piece I've been doing with my agents and didn't have a name for. I run a team of 6 AI agents. Each one has instruction files, memory, shared context. When something drifts, I tell the agent to go fix its own skill file. It reads the problem, rewrites the instruction, and updates the file. Already works. But I'm still the one noticing the drift. I'm still the trigger. Self-healing closes that gap. The agent watches its own execution. When a skill starts degrading, it rewrites the instruction, tests it, and rolls back if it made things worse. Observe → Inspect → Patch → Evaluate → Rollback. No human noticing. No human triggering. Context engineering got us to "agents that know things." Self-healing skills get us to "agents that stay good at things" without you babysitting.
Vasilije@tricalt

x.com/i/article/2032…

English
27
17
220
39.5K
Aakash Gupta
Aakash Gupta@aakashgupta·
Karpathy just mass-distributed a product without writing any code for it. The gist is one page of markdown describing how to build an LLM-maintained personal wiki. Raw docs go into a folder, an LLM "compiles" them into 100+ interlinked articles, and the whole thing self-heals through automated linting passes. His version runs on Obsidian with custom scripts and 400K words of compiled research. He calls it an "idea file." One page of markdown. Your agent reads it and builds the entire system customized to your stack in an afternoon. His Obsidian setup is probably thousands of lines of hacky Python. You'll never see any of it. A one-page spec and a capable agent gets you 90% of the way there, tuned for your own tools and workflow. The economics of this are wild. In the old model, someone shares a repo and maybe 2% of people actually clone it, install the dependencies, and get it running. Idea files flip that ratio. The agent handles all the translation work between "what Karpathy built" and "what works on your machine." The conversion rate from "saw a cool project" to "have my own version running" collapses from days to hours. Karpathy is basically saying the era of "here's my repo, go figure out the dependencies" is being replaced by "here's what I built and why, go tell your agent to make you one." Software distribution is becoming a game of telephone where every recipient gets a better version than the original.
Andrej Karpathy@karpathy

Wow, this tweet went very viral! I wanted share a possibly slightly improved version of the tweet in an "idea file". The idea of the idea file is that in this era of LLM agents, there is less of a point/need of sharing the specific code/app, you just share the idea, then the other person's agent customizes & builds it for your specific needs. So here's the idea in a gist format: gist.github.com/karpathy/442a6… You can give this to your agent and it can build you your own LLM wiki and guide you on how to use it etc. It's intentionally kept a little bit abstract/vague because there are so many directions to take this in. And ofc, people can adjust the idea or contribute their own in the Discussion which is cool.

English
25
46
574
102.1K
Babak Pahlavan
Babak Pahlavan@babakph·
Most AI apps are frozen the moment they ship. What if every app had an AI agent inside it that could modify the app itself? Not prompt customization, actual code-level mutation. That's the difference between a tool and a living product.
English
1
0
1
34
Ron Kurti
Ron Kurti@RonKurti·
What I want from my product in 2026. I want it to be alive. I want it to fix itself. Report to me. Run experiments. Self-heal. I want to design my web app in my web app. I want to collaborate with it on ideas. I want it to understand my users and surface the insights. I want it to notice when patterns shift and tell me before I have to ask. I want it to understand itself. I don't care about the stack. Cloud agents, local agents, multi-agents... I don't care what harness this runs on. I don't care how it chooses to self-organizes its own context. I don't want to engage with any of that. In the same way I stopped thinking about hosting, I want to stop thinking about LLM infrastructure. I just want to engage with my product. And I want my product to be alive. (Tagging @danshipper and @karpathy bc i think they might agree)
Ron Kurti tweet media
English
0
0
0
15
David Marcus
David Marcus@davidmarcus·
Not experimenting with @openclaw to the max extent for “security reasons” is like not connecting to the internet in the 90s.
English
40
13
236
29.1K
kevin
kevin@kleffew94·
I spun up EmDash via npm @Replit. It feels like a modern, Astro-based CMS, with built-in x402 support for per-page pricing and bot-aware monetization. Was also really fun to use Replit Canvas to edit the design natively. It almost feels like working on a live Figma board.
kevin tweet mediakevin tweet mediakevin tweet media
San Francisco, CA 🇺🇸 English
4
2
22
924
Roy Murphy
Roy Murphy@murphsicles·
⚠️Zeta v0.3.45 Advanced Metaprogramming: Code Transformation This release represents a major milestone in the Zeta compiler's evolution, bringing cutting-edge features and significant improvements to systems programming. github.com/murphsicles/ze…
English
1
0
1
230
Ron Kurti
Ron Kurti@RonKurti·
@nbbaier @penberg Its all auto-generated on the fly. The comments were an experiment in how to create another layer of information to help the reader understand the details.
English
1
0
0
3
David Aerne
David Aerne@meodai·
Heerich.js 0.10.0: finally a proper isometric view! Not faked with oblique math, but a real orthographic projection rotated ~35.264° like it should be.
English
7
7
100
5.2K
How To AI
How To AI@HowToAI_·
🚨 BREAKING: Vector databases for AI memory just got replaced by MP4 files. Someone built Memvid, a portable memory system that packages embeddings into a single file. It stores millions of text chunks using video encoding logic for sub-millisecond retrieval. → Replace expensive vector databases with single file. → Lightning-fast semantic search without a server. → Portable, versioned, and crash-safe AI memory. 100% open source.
How To AI tweet media
English
65
151
1.3K
90.9K
Jainam Parmar
Jainam Parmar@aiwithjainam·
Genuinely didn't expect this to exist yet. Someone built a text-to-CAD tool that runs completely in your browser. you describe what you want in plain English, upload a reference image if you have one, and it generates a real 3D model with interactive sliders to adjust every dimension. It exports as STL or SCAD so you can take it straight to a 3D printer. the whole thing runs on WebAssembly so there's nothing to install. Opensource. GPL-3.0 License. (Get the link the comments)
Jainam Parmar tweet media
English
7
63
492
24K
Nav Toor
Nav Toor@heynavtoor·
🚨 Someone reverse-engineered the design systems of Apple, Spotify, Airbnb, and 30+ billion-dollar companies. Packed each one into a single file. Free. It's called Awesome Design MD. Drop one file into your project. Your AI agent builds UI that looks like Spotify. Or Apple. Or Airbnb. Instantly. Not screenshots. Not Figma links. A single DESIGN .md file that captures every color, font, spacing value, button style, and layout pattern from a real website. In a format AI agents read and reproduce. Here's the difference: Tell Claude Code "build me a landing page" and it gives you generic UI. Tell Claude Code "build me a landing page" with Spotify's DESIGN .md in your project and it gives you Spotify. Here's what's inside: → Apple. Premium white space, SF Pro typography, cinematic imagery. → Spotify. Vibrant green on dark, bold type, album-art-driven layout. → Airbnb. Warm coral accent, photography-driven, rounded UI. → Linear. Ultra-minimal, precise spacing, purple accent. → SpaceX. Stark black and white, full-bleed imagery, futuristic. → BMW. Dark premium surfaces, precise German engineering aesthetic. → NVIDIA. Green-black energy, technical power aesthetic. → Uber. Bold black and white, tight type, urban energy. → Sentry, PostHog, Raycast, Cursor, ElevenLabs, and 20+ more. Here's how to use it: → Pick a design system from the collection → Copy the DESIGN .md file into your project root → Tell your AI agent to use it → Get UI that matches the design language of a billion-dollar company That's it. One file. Your AI agent now has the design taste of a $200/hour design consultant. Designers charge $5,000+ for a custom design system. Companies spend $50,000+ building one from scratch. This is free. 31 design systems. Copy. Paste. Ship beautiful UI. Works with Claude Code, Cursor, Codex, and any AI coding agent that reads project files. 100% Open Source. MIT License.
Nav Toor tweet media
English
45
323
3.4K
320.1K
bashbunni
bashbunni@sudobunni·
HE HAS RETURNED TO OPEN SOURCE DEVELOPMENT LET'S FREAKING GOOOOO github.com/maaslalani/she… Great person to follow if you're looking for project inspo. He's always building fun stuff.
English
11
27
466
21.6K
Andrew Farah
Andrew Farah@andrewfarah·
sharing my first open source project a CLI for downloading and syncing your X bookmarks locally so your agent can access them. it's free › npm install -g fieldtheory › login to your X account in a chrome tab › ft sync (done!) bonus: › ft viz › ft classify
English
131
92
1.7K
83.6K
Charles Packer
Charles Packer@charlespacker·
All "memory plugin" companies are doomed to fail - memory *is* the harness. This was clear going as far back as MemGPT (2023) - MemGPT is a harness that provides affordances to the LLM to edit memory pinned to its context window, not some external memory plugin.
Sarah Wooders@sarahwooders

x.com/i/article/2036…

English
2
0
34
6.9K