Austin

Thanks to @claudeai I made this whole coding app last weekend. This is my vision of future coding interface. #WIP

@om_patel5 Inference is not free, they are getting something from you

SOMEONE MADE IT SO YOU CAN CODE WITH 174 AI MODELS FROM 23 PROVIDERS FOR FREE one npm package and one install: "npm i -g free-coding-models" switch between any model instantly. benchmark them against each other in real time to see which one actually codes best for your use case. no subscriptions. no API keys. no cost. 174 models. 23 providers, all free, all from your terminal. if you're tired of paying for 3 different AI subscriptions just to compare which model codes better, this is it

@steipete @work_____ U would say some shit like that now. Sellout

@work_____ OpenClaw was mostly built by codex.

Bros openclaw without Claude is shit. I can tell you bc for 2 months I’ve tried OpenAI $200 and Claude $200 subs. OpenAI is the bitch of Claude. OpenAI is good to do some cleaning and smol cooking and that’s it. Important vibe coding is for Claude and Claude only.

@philkellr I have tried openclaw 10 different times. It’s shit. Can’t be trusted with anything serious. And vibe coding with it costs like 20x more token than just using Claude. Claude is great but expensive and they are turning into control freaks. OpenAI was my gateway into AI. Grokgood

I'm tired of OpenClaw Every 2-3 days I have a major moment which I show to my friends: "look what AI agents can do", but then the other 90% is pure frustration and me cursing at my own AI agent for which I spent hours choosing a beautiful name and profile picture. I was fun at start: I added telegram, added voice input, then adding skills even with just voice prompting was a bliss. Then dementia hit. Facts from 48+ hours ago were forgotten. I installed a 3 level memory system. It felt like a huge hack, it barely works, I encounter bugs every day which I'm fixing. Then it breaks with every update. Not all of it, but little things. The WhatsApp integration is just insanity. After putting in every markdown file that it shouldn't reply to my friends (IN CAPITAL LETTERS) it happily started to chat with my wife and my goddaughter. And from today even claude subscription stops working. I feel like a failure! I see all the success stories left and right, YT vids and blog posts "I got it to work and here's what the AI agent does for me", and for me I'm still spending 5x the time fixing my agent than just doing the stuff by hand 🤷

@Abomination81 @TraderRomeo23 If yours makes more money than it loses, it’s 100x cooler than mine lol

@agentic_austin @TraderRomeo23 I have one, but it doesn't look that cool.... yet lmao

Just finished creating a new frontend for my bots. Love to see the ones you guys created (placeholder data, just polishing it up)

@Abomination81 here is another polymarket trading bot i made, also only profitable in paper mode. Spread harvester strategy. Also open source at github.com/austindixson/p…

@agentic_austin How awesome looking. Where is the globe/weather graphics from?! That is amazing looking

@Abomination81 Claude got me started with a 3d globe, and then I was able to overlay the wind/temp overlay from github.com/cambecc/earth

@TraderRomeo23 @Abomination81 weather markets move too slow for me lol it was fun to build though. anyone that wants to build a wether trading app that has more patience than me has a good jumping off point!

@agentic_austin @Abomination81 May I know what's the reason for giving up such a sophisticated project.

@Abomination81 I made a dope polymarket weather trading dashboard, gets weather data from airports. gave up on this project but I made it opensource. Give it a star github.com/austindixson/w…

@Abomination81 update it sucks at everything

Google Just released TimesFM. Its a predictive ML that you don't have to train. It's open source and trained on 100B data points. Zero Shot forecasting. Downloading it now. You guys should too. Super curious to see if anyone can find a market this nukes. I'm going to start with crypto!

@sudoingX My experience with nemotron has been bad. Seems slow and not that smart imo but I was running through openrouter, maybe that’s the difference

nvidia dropped a 120 billion parameter model with an architecture nobody has stress tested yet. so i did what any sane person would do. loaded the full thing unquantized on 2x H200 NVL and gave it one job. build me a GPU marketplace from scratch. coded 769 lines in a single file from one prompt on hermes agent. one shot. served it successfully to localhost. Nemotron Super 120B-A12B. Mamba-2 + MoE + Attention hybrid with LatentMoE routing. 12B active params out of 120B total. 60.2 tok/s at BF16 unquantized across 287GB of VRAM. the model understands structure. hero sections, marketplace listings, template pages. it has weight to it. production feel. there are flaws. boxed widths, text overlapping, no particles. but if the prompt was more explicit it would have nailed those too. this model thinks before it writes. this was the warm up. single file challenge. today it faces the real test. octopus invaders. multi-file autonomous build with hermes agent. a game i open sourced so anyone can test their model against the same challenge. first built on a single 3090. now these flagships try it at datacenter scale. watch the video to the end for the full UI walkthrough.

@BentoBoiNFT I created over 50 openclaw skills just to realize i was trying to make openclaw more like claude code.

OpenClaw is amazing when it works, but it takes many hours fixing the problems and issues that pop up Could be a skill issue and I just need to work on it more. I’ll still be using it in my workflow, but I’m looking at other options too right now.

OpenClaw is amazing as a concept but I am getting genuinely sick and tired of at least one thing breaking every single day

Also its opensource, anyone can download and contribute to it. Free and yours forever. github.com/austindixson/a…

Thanks to @claudeai I made this whole coding app last weekend. This is my vision of future coding interface. #WIP

Works pretty good for a “partially assembled” engine :)

🚀 codemap-organizer v2. Now 21 languages ✨ What's New: • Swift support • Rust rewrite + dependency extraction • Kotlin, C#, Dart, Ruby, PHP, C++, PowerShell, YAML support 🧠 Under the Hood: • Semantic keyword extraction (finds exports, not paths) • Bidirectional dependency tracking (imports + imported by) • Smart purpose inference (analyzes code structure) • Incremental mode (--incremental flag) 📊 Results: • 96-98% token reduction across TypeScript, Swift, Rust • +32% F1 score (precision: 0.48 → 0.67) • Zero false positives in Swift test 🔗 github.com/RuneweaverStud… AI-friendly code navigation without the bloat.

@karpathy @elonmusk

Just shipped a codemap-organizer Adds AI-readable headers to code files Results: • 96% token reduction (4,333 → 150 per search) • +32% accuracy improvement • Works with TS, Py, Go, Rust, C++ AI no longer needs to read entire files to navigate your codebase GitHub: github.com/RuneweaverStud… #ClaudeCode #AI

@elonmusk Save Seattle please

This is insane