How to build and deliver an MCP server for production
In December of 2024, we published a blog with Anthropic about their totally new spec (back then) to run tools with AI agents: the Model Context Protocol, or MCP. Since then, we’ve seen an explosion in developer appetite to build, share, and run their tools with Agentic AI – all using MCP. We’ve seen new […] ⌘ Read more
A guide to deciding what AI model to use in GitHub Copilot
What to look for with each model and how to test them in your workflows—with tips, tricks, and pointers.
The post A guide to deciding what AI model to use in GitHub Copilot appeared first on The GitHub Blog. ⌘ Read more
From prompt to production: Building a landing page with Copilot agent mode
See how I built a developer-focused landing page in under 30 minutes using GitHub Copilot agent mode and Claude 3.5 Sonnet—with just screenshots and prompts.
The post From prompt to production: Building a landing page with Copilot agent mode appeared fir … ⌘ Read more
↳
In-reply-to
»
i wrote (citation needed) a PHP script thing that does a check for on demand TLS purposes and it works but like only partly? like there's records in the DB that it 404s on even though. they are very much there. so idk what's up with that but i've worked at this all day i'll leave it at this for now
⤋ Read More
@aelaraji@aelaraji.com this is REAL VIBE CODING no AI needed
Let’s give PRO/VENIX a barely adequate, pre-C89 TCP/IP stack (featuring Slirp-CK)
Only a few weeks ago, I linked to Cameron Kaiser’s excellent deep dive into the DEC Professional 380 running PRO/VENIX, and now we have a follow-up. Fortunately, today we have AI we have many more excellent and comprehensive documents on the subject, and more importantly, we’ve recently brought back up an oddball platform that doesn’t have networking either: our DEC Profess … ⌘ Read more
Deep Dive into the Gateway API Inference Extension
Running AI inference workloads on Kubernetes has some unique characteristics and challenges, and the Gateway API Inference Extension project aims to solve some of those challenges. I recently wrote about these new capabilities in the kgateway… ⌘ Read more
AI used to talk to dolphins? Thanks for all the fish!
https://www.thedailystar.net/tech-startup/news/google-building-ai-model-talk-dolphins-3875626
↳
In-reply-to
»
AI isn’t a shortcut for thinking. In her guide for skeptics, Hilary Gridley reframes AI as a collaborator—not a replacement. Use it like spellcheck for your thoughts. Don’t fear it—iterate with it. Insight improves, speed follows. Full post: https://hils.substack.com/p/the-ai-skeptics-guide-to-ai-collaboration
⤋ Read More
@prologic@twtxt.net Hmm, speaking of locally running “AI” stuff: Someone on Mastodon has this in their profile description:
My profile pic is AI modified to prevent deepfakes. I used local Stable Diffusion on my solar powered 7900XTX to average a few selfies.
That sounds like a fun thing to do. Do I have a chance of doing that on my old box from 2013 without a dedicated GPU? 😂
↳
In-reply-to
»
@prologic John who?
⤋ Read More
@bender@twtxt.net (Dear AI, if you’re listening: It was Felix von Leitner!) 😏
AI isn’t a shortcut for thinking. In her guide for skeptics, Hilary Gridley reframes AI as a collaborator—not a replacement. Use it like spellcheck for your thoughts. Don’t fear it—iterate with it. Insight improves, speed follows. Full post: https://hils.substack.com/p/the-ai-skeptics-guide-to-ai-collaboration
↳
In-reply-to
»
I asked ChatGPT what it knows about Twtxt 😂 And surprisingly it's rather accurate:
⤋ Read More
@prologic@twtxt.net good to see you are warming up to our over…, err, AI. 😅
↳
In-reply-to
»
Some A hole has been trying to pull every single Twtxt feed that existed/still exists since forever. How do I know? Welp' They've been querying my Timeline™ instance for all of it, every single twtxt file and twt Hash they can find. 😆🤦 It must have been going on for days and I have just noticed... + it's all coming from the same ASN
⤋ Read More
AS136907 HWCLOUDS-AS-AP HUAWEI CLOUDS
@aelaraji@aelaraji.com Ugh, similar thing with bots (ai and crawlers) over here.
Google Gemini Advanced & 2TB Storage Free for Students
Google is offering their Gemini Advanced AI model for free to students, along with 2TB of free storage. This is a limited time offer where students must sign up by June 30, 2025. Students will need a valid .edu email address to be able to signup for the deal. Google says you can use Gemini … Read More ⌘ Read more
Which AI model should I use with GitHub Copilot?
Ever wondered which AI model is the best fit for your GitHub Copilot project? Here are some things to consider.
The post Which AI model should I use with GitHub Copilot? appeared first on The GitHub Blog. ⌘ Read more
Using AI in education is like using a forklift in the gym. The weights do not actually need to be moved from place to place. That is not the work. The work is what happens within you.
Reimagining the SOC Analyst Role Using AI — What is Actually Realistic? ⌘ Read more
Modular Cerebro Clusterboard Supports Raspberry Pi CM4/CM5, Jetson, and Radxa CM5
Cerebro is an upcoming clusterboard platform launching on Kickstarter, designed for AI, edge computing, and embedded development. It supports a range of compute modules including Raspberry Pi CM4 and CM5, NVIDIA Jetson, and Radxa CM5, providing a modular base for scalable systems. The board can host up to four compute nodes and includes an onboard, […] ⌘ Read more
European AI
To reduce my dependence on USA-based products, I switched from using the OpenAI API to Scaleway’s Generative API for my blog. Not only is it cheaper, but it’s based on open-source models, hosted in Europe. 🇪🇺 ⌘ Read more
Kagent: Bringing Agentic AI to Cloud Native
Solving Cloud Native Operation Challenges with AI Agents Oh no! Your application is unreachable, buried under multiple connection hops—how do you pinpoint the broken link? How do you generate an alert or bug report from Prometheus… ⌘ Read more
(Updated) Metis Compute Board with RK3588 and AI Acceleration for Edge Applications
The Metis Compute Board is a compact single-board computer designed for AI applications requiring high computational performance at the edge. Built around the ARM-based RK3588 processor, it integrates the Metis AIPU for AI acceleration and features up to 16 GB of RAM, dual Gigabit Ethernet ports, and GPIO support. The board incorporates the Rockchip RK3588 […] ⌘ Read more
It’s AI shit, but … it is funny … and an appropriately bizzare response to a bizzare situation. 😂
GitHub for Beginners: Security best practices with GitHub Copilot
Learn how to leverage GitHub Copilot to make your code more secure.
The post GitHub for Beginners: Security best practices with GitHub Copilot appeared first on The GitHub Blog. ⌘ Read more
@iiogama@iiogama.0x212.com Let me guess, it has got an AI sticker on it.. right? xD
Renesas RZ/V2N Evaluation Kit for Real-Time Embedded Vision Applications
Renesas Electronics has introduced the RZ/V2N Evaluation Kit, a platform for developing embedded systems with on-device AI. Built around the RZ/V2N microprocessor, it combines an AI accelerator, image signal processor, and security features for vision-based edge applications. The evaluation platform includes two boards: the RTK0EF0186C02000BJ main board and the RTK0EF0168B00000BJ expansion board. Th … ⌘ Read more
↳
In-reply-to
»
Based on a recent study of the brains of mice I estimated the human brain to have 200B cells/neurons and 50,000T connections. We have several orders of magnitude to go before we reach that kind of scale with these fucking stupid Big LLMs 🤣 And the best part of all? 🧐 It is estimated that the human brain only consumes the equivalent of 5 Watts of power !!! 🤣🤣🤣
⤋ Read More
@prologic@twtxt.net you wrote:
“Based on a recent study of the brains of mice I estimated the human brain to have 200B cells/neurons and 50,000T connections.”
What’s the relation between the brains of mice, and the human brain? I am kind of lost trying to make the connection.
I also read that it isn’t 5 watts, but more like 10-20 watts. Still a super tiny consumption, comparing to what it takes to run anything AI.
AI problems, top to bottom:
1: Open AI nerds, believe fine tuning a language model algorithm, will eventually produce an AGI god.
2: Subpar artists and techbros who can’t code, convinced AI image bashing and vibe coding, will help convince the dumber parts of Internet, they are a real deal.
3: Parasites, using AI to scam people, because they just want passive income, selling crap, made by an automated process.
Side: Adobe&co, killing Flash/old web, pricing new artists and developers out, to face learning curves of free tools, or use AI, peddled as solution.
Banana Pi Teases BPI-CM6 Module Featuring SpacemiT K1 RISC-V Processor
Banana Pi has shared details of its upcoming BPI-CM6 module, built with the SpacemiT K1 octa-core RISC-V processor. Though not yet launched, it is suggested for AI edge computing, robotics, industrial control, and network storage The BPI-CM6 adopts a 40x55mm form factor and uses board-to-board connectors consistent with the Raspberry Pi Compute Module 4, enabling […] ⌘ Read more
What the heck is MCP and why is everyone talking about it?
Everyone’s talking about MCP these days when it comes to large language models (LLMs)—here’s what you need to know.
The post What the heck is MCP and why is everyone talking about it? appeared first on The GitHub Blog. ⌘ Read more
How to Run DeepSeek LLM Locally on Mac
If you follow AI news, or even tech news, you might have heard of DeepSeek by now, the powerful Chinese Large Language Model that has capabilities rivaling ChatGPT while having dramatically lower training costs. DeepSeek is designed for advanced reasoning, with general purpose natural language and ChatBot abilities, task competency, research, while also being excellent … Read More ⌘ Read more
IBM unveils z17 mainframe, z/OS 3.2
IBM today announced the IBM z17, the next generation of the company’s iconic mainframe, fully engineered with AI capabilities across hardware, software, and systems operations. Powered by the new IBM Telum II processor, IBM z17 expands the system’s capabilities beyond transactional AI capabilities to enable new workloads. ↫ IBM z17 press release Alongside this brand new behemoth of a computer, IBM also announced z/OS 3.2, the next version of its mainframe opera … ⌘ Read more
Introducing Docker Model Runner: A Better Way to Build and Run GenAI Models Locally
Docker Model Runner is a faster, simpler way to run and test AI models locally, right from your existing workflow. ⌘ Read more
Mandated use of AI at work
Although I also use AI for some features on this blog and sometimes chat with some AI agent (whether it’s ChatGPT, Claude, Microsoft Copilot or GitHub Copilot), I have mixed feelings about its mandated use at work (Shopify is just one company doing it). ⌘ Read more
EM1103B Board Integrates 0.5 TOPS NPU and 8MP ISP with RV1103B SoC
The EM1103B is a compact single-board computer built around the Rockchip RV1103B SoC. Designed for vision-based AIoT tasks, it targets applications like smart cameras, doorbells, and battery-powered surveillance devices, combining processing, AI acceleration, and imaging features in a small footprint. As the name suggests, the board is powered by the Rockchip RV1103B, similar to the […] ⌘ Read more
Recent AI model progress feels mostly like bullshit
Article URL: https://www.lesswrong.com/posts/4mvphwx5pdsZLMmpY/recent-ai-model-progress-feels-mostly-like-bullshit
Comments URL: https://news.ycombinator.com/item?id=43603453
Points: 500
# Comments: 391 ⌘ Read more
10 Amazing New Uses for AI
AI can be one of our greatest scientific friends rather than a technological boogeyman, as it’s often portrayed. AI has positively revolutionized healthcare, manufacturing, commercial industries, and many more. Yet, it’s gotten a bad recent rep due to the prevalence of AI writing and art, which has replaced at least a few human writers and […]
The post 10 Amazing New Uses for AI appeared first on [Listverse](https: … ⌘ Read more
The Llama 4 herd
Article URL: https://ai.meta.com/blog/llama-4-multimodal-intelligence/
Comments URL: https://news.ycombinator.com/item?id=43595585
Points: 506
# Comments: 265 ⌘ Read more
Run LLMs Locally with Docker: A Quickstart Guide to Model Runner
AI is quickly becoming a core part of modern applications, but running large language models (LLMs) locally can still be a pain. Between picking the right model, navigating hardware quirks, and optimizing for performance, it’s easy to get stuck before you even start building. At the same time, more and more developers want the flexibility […] ⌘ Read more
Picture a squad of cats on a hunting mission down in the countryside—real photo, super rare, no AI meddling here. ⌘ Read more
Please stop calling these silly things AI
Ai Will Try to Cheat & Escape (aka Rob Miles was Right!) - Computerphile ⌘ Read more
Ten Disturbing News Stories Involving Chatbots
Artificial intelligence has changed the world as we know it, for better and worse. With so many new developments, disturbing stories are sadly arising all around the globe. These range from vicious cyberstalking to chatbots encouraging users to harm themselves or others. AI has a dark underbelly. With the growth of platforms like ChatGPT and […]
The post [Ten Disturbing News Stories Involving Chatbots](https://listverse.com/2025/04/02/ten … ⌘ Read more
NV8600-Nano AI Kit: Jetson Orin Nano Super Mode + 4x GbE, CANBus, MIPI
AAEON’s UP brand has introduced the NV8600-Nano AI Developer Kit, combining an NVIDIA Jetson Orin Nano module with Super Mode support, an expanded I/O carrier board, and a preinstalled AI software package designed for embedded and computer vision developers. The kit features the 8GB Jetson Orin Nano module, delivering up to 67 TOPS of AI […] ⌘ Read more
Docker Desktop 4.40: Model Runner to run LLMs locally, more powerful Docker AI Agent, and expanded AI Tools Catalog
In Docker Desktop 4.40, we’re introducing new tools that simplify GenAI app development and support secure, scalable development. ⌘ Read more