Justin

Legacy shops build grit and resourcefulness. High-tech shops build systems thinking and scalability

@DabsMalone Lot more metal 4sure

Its hard to imagine what the world will look like in 5 years

All I know is this software stuff drains the life out of me. That’s my biggest motivation for putting agents to work.

@willreil This is way

Good habits will set you free. Build a new identity one small step at a time.

@yacineMTB It gives me the most foul analogies and keeps referring to me by my professional title

man. gemini is so bad

Are products the only things that exist in the real world? Services just support products?

@socialwithaayan It’s happening

🚨 BREAKING: Someone just built the exact tool Andrej Karpathy said someone should build. 48 hours after Karpathy posted his LLM Knowledge Bases workflow, this showed up on GitHub. It's called Graphify. One command. Any folder. Full knowledge graph. Point it at any folder. Run /graphify inside Claude Code. Walk away. Here is what comes out the other side: -> A navigable knowledge graph of everything in that folder -> An Obsidian vault with backlinked articles -> A wiki that starts at index. md and maps every concept cluster -> Plain English Q&A over your entire codebase or research folder You can ask it things like: "What calls this function?" "What connects these two concepts?" "What are the most important nodes in this project?" No vector database. No setup. No config files. The token efficiency number is what got me: 71.5x fewer tokens per query compared to reading raw files. That is not a small improvement. That is a completely different paradigm for how AI agents reason over large codebases. What it supports: -> Code in 13 programming languages -> PDFs -> Images via Claude Vision -> Markdown files Install in one line: pip install graphify && graphify install Then type /graphify in Claude Code and point it at anything. Karpathy asked. Someone delivered in 48 hours. That is the pace of 2026. Open Source. Free.

@aegrabs Sometimes big nozzle is right

does carbon fibre filament really abrade nozzles that quickly or is that just a myth made up by big nozzle

@googlegemma @Scobleizer Can’t even tell me the weather smh

@Joestar_sann It’s certainly a hype cycle but local hybrid agents seems to be the long-term direction when you look past all of it. You can’t have a robust model with the Gemma 4 models alone. It’s all about tokens/s/$

so let me get this straight all of ai twitter was telling people to buy a mac mini to run openclaw, which is literally just a framework, an orchestration layer that sends api requests to actual ai models. something you can run on a $5/month vps. which is exactly what i do btw but when google drops gemma 4, an actual large language model that you can run and fine-tune locally on that same mac mini, with no api costs, no subscriptions, no third party dependencies, completely yours under apache 2.0 the ai community is silent you were buying $800 hardware to run a wrapper but ignoring the actual ai model that would justify that hardware this tells you everything you need to know about the average iq of ai twitter

@yosoymario91 @signulll The niche ones will have the real value

@signulll I doubt there's even one, all the use cases ive seen are either extremely niche or just some nonsense like getting a Wikipedia summary at the morning on what happened in a particular day

i’m curious what are your top 5 use cases for openclaw that you cannot live without?

@elnomic @signulll Cope

@signulll literally nothing, claude cowork has a much more controlled environment

@darthcap1tal @signulll Hahahahaha

@signulll This, of course

@bridgemindai @matthewmillerai So you will still need API models for higher reasoning and complex execution right? I have a tiny setup with my 5060Ti 16GB and already have issues with simple tool calling like web search.

OpenClaw running on my NVIDIA DGX Spark. No Anthropic servers. No rate limits. No subscription. Claude Code rate limited me. Anthropic cut off OpenClaw from subscription limits. So I bought my own hardware. Now I'm running OpenClaw locally on a Grace Blackwell chip. The same tool Anthropic tried to throttle, running on infrastructure they can't touch.

マイクロなマイクロメーターつくりました

@dbdanieljnr @SavegeorgeG This more for industrial design right?

@make2spec @SavegeorgeG Speedy prototyping. say you are discussing product feature with a client, and you already prepared a keyword capturing agents through speech to text, then with great tooling and skills, one can have a client interact with a prototype in the first discovery brainstorming session.

That I can create a part in OpenSCAD, place it along others in FreeCAD, then export it, and apply materials in Blender, all from terminal, is NUTS!!! Dunno if I should show my boss or not!

@matterasmachine The cell phones are just a tool

@make2spec You cannot align civilisation with nature - no cell phones in the wild.

"Ego" is good if you are right.

Finally got my local AI running with Gemma 4. My benchmark will only be how much it can do for me day to day.

@willreil I'm always way too conservative with margins. If no one else is in the market you really have a choice to price it what it's worth or what people will pay for it

If you aren’t fighting for every cent, you aren’t going to make it. A cent turns into a dollar. Before you know it, your margins have been eaten to the bone by complacency.