A few months back, inspired by a great session by @edjgeek , we made changes to an AWS step function reducing its cost from $450 per invocation to around $1. If you're intrigued, check the first part of a two-part article I wrote below
medium.com/chronicles-of-…
Introducing Claude Sonnet 4.5—the best coding model in the world.
It's the strongest model for building complex agents. It's the best model at using computers. And it shows substantial gains on tests of reasoning and math.
Am I weird if I think this isn't truly solving a problem, and will instead just drive prices higher so that merchants will cover the "small fee" they are charged by OpenAI for each transaction?
Are we really that lazy that we can't check out/buy things ourselves anymore?
Am I weird if I think this isn't truly solving a problem, and will instead just drive prices higher so that merchants will cover the "small fee" they are charged by OpenAI for each transaction?
Are we really that lazy that we can't check out/buy things ourselves anymore?
We’re introducing the Agentic Commerce Protocol, an open standard co-developed with @stripe that enables programmatic commerce flows between users, AI agents, and businesses.
We’re introducing the Agentic Commerce Protocol, an open standard co-developed with @stripe that enables programmatic commerce flows between users, AI agents, and businesses.
AI agents are now capable of doing real, if bounded, work. But that work can be very valuable. For example, the new Claude Sonnet 4.5 was able to replicate published economics research from data files & the paper.
We need to figure out what to do with it: oneusefulthing.org/p/real-ai-agen…
The jump from "agents are nowhere close to working" to "okay, narrow agents for research and coding work pretty well" to (very recently) "general purpose agents are actually useful for a range of tasks" has been quick enough (less than a year) so that most people have missed it.
I have been waiting for this for a long time, as I was tired of the hacky or wasteful approaches I had to resort to for better handling browser-side issues dev-ing
Now we have Cursor opening up a Chrome window and doing the testing, troubleshooting, and fixing 😀
I have been waiting for this for a long time, as I was tired of the hacky or wasteful approaches I had to resort to for better handling browser-side issues dev-ing
Now we have Cursor opening up a Chrome window and doing the testing, troubleshooting, and fixing 😀
Never thought of this, but it makes SO MUCH SENSE!
Which makes me wonder why it is the norm to disallow seeing the system prompt? Jailbraking and malevolent users?
Never thought of this, but it makes SO MUCH SENSE!
Which makes me wonder why it is the norm to disallow seeing the system prompt? Jailbraking and malevolent users?
My favorite AI workflow lately is my thought-to-post pipeline.
I just go on walks, have a good content idea, ramble it, and have an optimized post in my writing style without typing.
It's super simple:
1. Download an AI-powered voice dictation app to your phone (I use Wispr…
Today we’re introducing GDPval, a new evaluation that measures AI on real-world, economically valuable tasks.
Evals ground progress in evidence instead of speculation and help track how AI improves at the kind of work that matters most.
openai.com/index/gdpval-v0
Most people ask if AI will replace their job. Wrong question. The real one: how fast will your workflows get re-engineered around it, with or without you
WEEK 6 PROJECT REVEAL: I just built an AI Content Authenticity Detector that can spot AI-generated text, images, and audio with 90%+ accuracy.
It uses GPT-4, ensemble models, and runs in real-time through a browser extension.
Here's how it works and why it matters... 🧵
Totally makes sense, I like that labs are trying to imitate person-like conditions for LLMs that are designed to do person-like tasks.
Like in this relatively recent study
arxiv.org/abs/2508.15510
Totally makes sense, I like that labs are trying to imitate person-like conditions for LLMs that are designed to do person-like tasks.
Like in this relatively recent study
arxiv.org/abs/2508.15510
Hidden inefficiencies cost more than hallucinations.
Our anti-pattern system spotted passive learning loops in education with 95% accuracy, saving ~90% compute.
That’s where AI creates ROI: workflow, not just answers.
trilogyai.substack.com/p/anti-pattern…
25 Followers 210 FollowingAnthropomorphic Creative Force
AI Engineer
PhD in Statistical Physics, CBPF (Brazilian Center for Physical Research)
Musician
97 Followers 938 Following20+ yrs in building and hacking software | Applied AI/ML | Full Stack Dev | 2x Founder | IIT Kharagpur alum | Building @EPHOR_ai @ Trilogy
66 Followers 147 Following🤖AI Enthusiast | 📊FinOps & BizOps Disruptor | Harnessing the power of AI to streamline and automate FinOps processes. #AIFirst #FinOpsDisruption
128 Followers 566 FollowingGenAI first tech leader:
* Transforming support and sales agent into cyborgs
* Simplifying SMSC and Communication products
* Cost Optimization guru
44 Followers 134 FollowingA finance leader aiming to unlock the power of AI to revolutionize finance. Join me on the journey to a smarter, faster, leaner finance org. 💰🤖 #FinanceAI
19 Followers 737 FollowingAI will improve human understanding and social productivity, solve energy and climate issues, cure health problems, and lead toward an age of abundance.
2K Followers 2K FollowingHead of Growth @ConsensusNLP (prev. @VoiceflowHQ) || I plan pop up raves around Toronto @ https://t.co/EUauOj7FI6 and cook on TikTok!
25 Followers 210 FollowingAnthropomorphic Creative Force
AI Engineer
PhD in Statistical Physics, CBPF (Brazilian Center for Physical Research)
Musician
10 Followers 18 FollowingAI-First Product Manager @Skyvera | Automating the human out of telecom | Building AI that actually works, not just demos | Free/Libre software FTW
97 Followers 938 Following20+ yrs in building and hacking software | Applied AI/ML | Full Stack Dev | 2x Founder | IIT Kharagpur alum | Building @EPHOR_ai @ Trilogy
66 Followers 147 Following🤖AI Enthusiast | 📊FinOps & BizOps Disruptor | Harnessing the power of AI to streamline and automate FinOps processes. #AIFirst #FinOpsDisruption
597 Followers 245 FollowingAI CEO Leading IgniteTech, Khoros and GFI Software's transformation to a fully AI-first organization. Join us: https://t.co/8NFeLEbixI
343K Followers 48 FollowingOne of the most widely read and trusted cybersecurity news sites, providing IT security professionals informed insights into the latest news and trends.
137K Followers 1 FollowingClaude is an AI assistant built by @anthropicai to be safe, accurate, and secure. Talk to Claude on https://t.co/ZhTwG8dz3D or download the app.
69 Followers 116 FollowingCloud Solutions Architect at @engineyard. Opensource advocate. Bearded. Father of girls. Tweets are my own and not the views of my employer.
6K Followers 1K Followingbuilding the post-IDE IDE at https://t.co/hDpglja33W - coined “context engineering”, prev @replicatedhq @SproutSocial - ai that works pod @ https://t.co/69BhaNtWfd
528K Followers 882 FollowingI run a portfolio of internet companies and host @startupideaspod. CEO: @latecheckoutplz we build companies like @ideabrowser, @meetLCA, @boringmarketer etc
69 Followers 126 FollowingHold on, AI is creating my bio. User of GenAI for... anything: work (product, engineering), products (eg. https://t.co/wwaVDa4DrP), interests & hobbies, learning.
128 Followers 566 FollowingGenAI first tech leader:
* Transforming support and sales agent into cyborgs
* Simplifying SMSC and Communication products
* Cost Optimization guru
63 Followers 29 FollowingAI engineering agency. We turn ideas into prod apps in 2 weeks. No upfront cost, pay only if we exceed expectations. Built for founders who move fast.
105 Followers 85 FollowingSVP of Eng at Trilogy. Former TA & start-up founder. 🚀
Tech chameleon: SAP, Spring, Node, AWS, React. 🌐
Exploring GenAI in development. 🤖
#GenAI #TechLead