Founder of ONE https://xitter.cf/t.co/yhLMVY8TMw & https://xitter.cf/t.co/RBJDLClHaJ Creating emergent AI. I love to build fast things with Claude Code, Astro, TypeDB, X402, SUI, Fetch AI
Dublin, Ireland
Joined November 2008
- Tweets2K
- Following1.4K
- Followers3.3K
- Likes974
Pinned Tweet
Did you know you can add Claude Code @claudeai from @AnthropicAI as a provider to Vercel's @vercel AI SDK @aisdk Watch how you can interact with your .claude/agents team with your browser
Tony O'Connell retweeted
Introducing SKY-001, the first Engineering World Model.
On CAD and BIM tasks, it beats GPT-6 Astra and every frontier model we tested:
-> 51-125x faster
- > 150-210x cheaper
-> Higher IOU than all frontier models
SKY-001 lets you dream a physical object before you build it.
18 months of work. Built by the team out of Maluuba's deep learning lab (acq. by @MSRResearch).
The first ‘AI General Engineer’ will be the backbone for Autonomous Manufacturing.
Tony O'Connell retweeted
Want to try GLM 5.2 in production but worried how it might change your product?
Don’t worry, we got you:
1. Install Inference Gateway (docs.inference.net)
2. Keep sending traffic to your current provider
3. Gateway automatically starts sorting through your live data using an RLM to generate evals for your app. This takes ~24 hours.
4. Gateway starts mirroring live traffic to GLM 5.2 to run evals. Traffic is only mirrored - you’re still using your old provider in prod.
5. Once evals look healthy, you get a Slack notification letting you know it’s safe to switch.
6. Switch model identifier in your code to “glm-5.2”
Congrats, you just saved 90% on your monthly token bill, and you own your LLM stack end to end.
Tony O'Connell retweeted
circle built x402 so AI agents pay for API calls, compute, and inter-agent services using USDC with zero human intervention. EIP-3009 means the agent never needs gas tokens, the service provider pays gas and the agent just signs a transfer authorization. $1.2b+ in AI crypto funding deployed last month and every single project needs programmatic payments to function. there's no x402 token to buy. the demand accrues to USDC float. every autonomous agent running 24/7 needs a pre-funded USDC balance just to operate. that's a permanent new source of stablecoin demand from non-human actors that doesn't exist in any model yet
Tony O'Connell retweeted
The thesis is simple: the future belongs to individuals who build compounding AI systems, not to individuals who use corporate-owned centralized AI tools.
I'm trying to build these in open source so you can have them for free. That's what GBrain is.
Tony O'Connell retweeted
Tencent has killed fine-tuning and RL with a $18 budget.
Right now, if you want an AI agent to become an expert at a specific, complex real-world task, you have to use Reinforcement Learning.
You let it try, fail, and update its internal parameters over and over again.
This is the exact optimization technique (GRPO) that DeepSeek used to build their massive reasoning models.
But there is a massive problem.
Updating model weights is insanely expensive. It requires massive GPU clusters. And worst of all, when you train a model to be highly specialized at one thing, it often "overfits" and forgets how to be good at everything else.
Tencent killed this bottleneck forever.. by building Training-Free GRPO.
Instead of spending thousands of dollars to permanently alter the AI's brain, they asked a simple question: What if we just distill the experience of learning, and inject it as a memory?
Here is how it works.
They run the AI through the exact same trial-and-error process. But instead of updating the weights, they extract the "semantic advantage"—the actual logic of why one answer was better than another.
They compress this winning logic into a "token prior”, a tiny package of high-quality experiential knowledge.
Then, they just attach that knowledge directly into the API call.
The results are staggering.
Tested on DeepSeek-V3, this method required only a few dozen training samples to turn the AI into a specialized expert in complex math and web searching.
It didn't just compete with models that were actually fine-tuned. It outperformed them.
Zero parameter updates. Zero expensive training runs. Zero base-model amnesia.
Tony O'Connell retweeted
Claude Code cannot read 300 files at once.
So someone built a system that lets it control NotebookLM from the terminal instead. The results are wild.
Here is the full workflow nobody is talking about:
The Setup
→ Claude Code connects to NotebookLM via a command line interface
→ Claude searches YouTube, finds relevant videos, uploads them as sources automatically
→ NotebookLM processes up to 300 sources simultaneously and returns cited, grounded answers
→ Everything syncs back into your Obsidian vault with passage-level citations you can click to verify
Why This Changes Research Forever
→ No more 20 browser tabs you never close
→ No more copy-pasting outputs into random notes
→ No more hallucinated answers with no sources to back them up
→ 60% of citations verified as strong matches in accuracy audits - answers are grounded in real data
What Claude Can Do From the Terminal
→ Search YouTube for relevant videos on any topic and rank by relevance
→ Create a new NotebookLM notebook and add 20 sources in parallel automatically
→ Ask questions and export cited answers directly into Obsidian with wikilinks
→ Set custom personas per notebook - concise, no filler, no preamble
→ Generate audio overviews and save them as MP3 files into your vault
→ Build mind maps, flashcard decks, and research dashboards from your sources
→ Search arXiv for academic papers and feed them directly into NotebookLM
→ Upload competitor blog posts, podcast episodes, PDFs, and your own vault notes
The Obsidian Output
→ Every answer arrives with clickable citations that link to the exact passage in the source video or article
→ Graph view shows connections between all 20 sources and the topics they share
→ Q&A log tracks every question asked and the grounded response received
→ Source dashboard shows citation frequency, topics extracted, and which questions each source answered
Use Cases Worth Building Today
→ Academic research with arXiv papers, full citation traceability
→ Competitor analysis from their YouTube channels and blog posts
→ Company knowledge base for onboarding, new employees ask NotebookLM instead of interrupting teammates
→ Podcast research, feed 4-hour Lex Fridman episodes and ask what's new in AI this week
→ Personal second brain, 300 daily notes uploaded and queryable in one notebook
Before this system existed you needed 20 tabs, hours of manual reading, and no guarantee the answers were real.
Now you type one prompt in the terminal and Claude does all of it for you.
The research stack of 2026 is not a browser. It is a terminal connected to everything
Tony O'Connell retweeted
x402 foundation officially formed yesterday under linux foundation. founding members: visa, mastercard, stripe, coinbase, circle, google, microsoft, amazon, shopify, cloudflare. average transaction value jumped from $1.60 to $34.50 in 30 days. 100m+ lifetime transactions. 99.9% market share of agentic payment rails. every major payment company on earth just organized around a single protocol for how AI agents move money. base captures 75% of x402 volume and is exploring a network token. the airdrop farmers are about to discover that building x402 integrations on base might be the most rational thing they've ever done
Tony O'Connell retweeted
I wanted to share a bunch of my favorite hidden and under-utilized features in Claude Code. I'll focus on the ones I use the most.
Here goes.
Tony O'Connell retweeted
My dear front-end developers (and anyone who’s interested in the future of interfaces):
I have crawled through depths of hell to bring you, for the foreseeable years, one of the more important foundational pieces of UI engineering (if not in implementation then certainly at least in concept):
Fast, accurate and comprehensive userland text measurement algorithm in pure TypeScript, usable for laying out entire web pages without CSS, bypassing DOM measurements and reflow
Tony O'Connell retweeted
🚨 Someone just open sourced a fully autonomous AI hacker and it's terrifying.
It's called Shannon.
Point it at your web app, and it doesn't just scan for vulnerabilities. It actually exploits them. Real injections. Real auth bypasses. Real database exfiltrations.
Not alerts. Not warnings. Actual working exploits with copy-paste proof-of-concepts.
Here's what this thing does autonomously:
→ Reads your entire source code to plan its attack
→ Maps every endpoint, API route, and auth mechanism
→ Runs Nmap, Subfinder, and WhatWeb for deep recon
→ Hunts for Injection, XSS, SSRF, and broken auth in parallel
→ Launches real browser-based exploits to prove each vulnerability
→ Generates a pentester-grade report with reproducible PoCs
Here's the wildest part:
It follows a strict "No Exploit, No Report" policy. If it can't actually break it, it doesn't report it. Zero false positives.
It pointed at OWASP Juice Shop and found 20+ critical vulnerabilities in a single run including complete auth bypass and full database exfiltration.
On the XBOW Benchmark (hint-free, source-aware), it scored 96.15%.
Your team ships code daily with Claude Code and Cursor. Your pentest happens once a year. That's 364 days of shipping blind.
Shannon closes that gap. One command. Fully autonomous.
The Red Team to your vibe-coding Blue team. Every Claude coder deserves their Shannon.
10.6K GitHub stars. 1.3K forks. Already trending.
100% Open Source. AGPL-3.0 License.
Tony O'Connell retweeted
🚨 Someone just solved the biggest bottleneck in AI agents. And it's a 12MB binary.
It's called Pinchtab. It gives any AI agent full browser control through a plain HTTP API.
Not locked to a framework. Not tied to an SDK. Any agent, any language, even curl.
No config. No setup. No dependencies. Just a single Go binary.
Here's why every existing solution is broken:
→ OpenClaw's browser? Only works inside OpenClaw
→ Playwright MCP? Framework-locked
→ Browser Use? Coupled to its own stack
Pinchtab is a standalone HTTP server. Your agent sends HTTP requests. That's it.
Here's what this thing does:
→ Launches and manages its own Chrome instances
→ Exposes an accessibility-first DOM tree with stable element refs
→ Click, type, scroll, navigate. All via simple HTTP calls
→ Built-in stealth mode that bypasses bot detection on major sites
→ Persistent sessions. Log in once, stays logged in across restarts
→ Multi-instance orchestration with a real-time dashboard
→ Works headless or headed (human does 2FA, agent takes over)
Here's the wildest part:
A full page snapshot costs ~800 tokens with Pinchtab's /text endpoint.
The same page via screenshots? ~10,000 tokens.
That's 13x cheaper. On a 50-page monitoring task, you're paying $0.01 instead of $0.30.
It even has smart diff mode. Only returns what changed since the last snapshot. Your agent stops re-reading the entire page every single call.
1.6K GitHub stars. 478 commits. 15 releases. Actively maintained.
100% Open Source. MIT License.
Tony O'Connell retweeted
Introducing cmux: the open-source terminal built for coding agents.
- Vertical tabs
- Blue rings around panes that need attention
- Built-in browser
- Based on Ghostty
When Claude Code needs you, the pane glows blue and the sidebar tells you why.
No Electron/Tauri. Just Swift/Appkit.
Tony O'Connell retweeted
Code Mode is all you need, very excited about this direction for MCP
blog.cloudflare.com/code-mod…
