el blanco

“People will come to love their oppression, to adore the technologies that undo their capacities to think.” — Aldous Huxley

@DavidOndrej1 Glorified CSV But it kinda works, right?

why the fuck is everyone fascinated with TOON ?!

@CryptoCloaks @CoQuiHODL @Cloudflare btc is to store value, not for rapid transactions.

@elblancoaudaz @CoQuiHODL @Cloudflare Solona is a shitcoin. Please stop.

Cloudflare introduces NET Dollar, a new U.S. dollar-backed stablecoin that will enable instant, secure transactions for the agentic web. cfl.re/4msrZWd

@CoQuiHODL @CryptoCloaks @Cloudflare una transaccion de btc tarda 10min, solana unos segundos y hay otros blockchains que son mas rapidos

@elblancoaudaz @CryptoCloaks @Cloudflare Maybe your brain is the one that's slow at catching up.

@CryptoCloaks @Cloudflare btc transactions are slow.

@Cloudflare Just use bitcoin you dummy's

@maxxmalist HORMOZI

I just trained a custom GPT on the following alex hormozi books: - 100M offers - 100M leads - 100M money models - the black books given to those who donated 200 books - + the playbooks, launch book and lost chapters like, rt + reply with "HORMOZI" and i'll send it straight to your DMs (must be following so i can dm)

@konstructivizm FYI: not visible standing on the Americas. timeanddate.com/eclipse/globe/…

On the night of September 7-8, 2025, nearly 7 billion people — about 85% of the world population — will witness a total lunar eclipse coinciding with a SuperMoon

@IterIntellectus Doubt it - this is a drop in PH's traffic, not in overall porn traffic. It's a re-routing of traffic.

ok, this may not be all bad

Cognitive Torque

@tyler_agg course

I just made an interactive course that gives everything you need to go from from beginner to confidently understanding how AI actually works in this course, you’ll learn: – how LLMs work under the hood – prompt engineering fundamentals – methods for providing context – how to avoid hallucinations comment “course” and I’ll DM you access (must be following)

@tyler_agg memory

I just made a walkthrough that shows you how to give an AI agent real memory - so it gets smarter the more you use it inside you'll learn: – what to track so it remembers user interactions – how to store and reuse that memory – how to personalize outputs based on past chats comment “memory” and I’ll DM you the video (must be following)

@tvanhens @thdxr I assume non-reasoning models give you more control, i sometimes get the feeling that reasoning models follow a certain path of what to do; but non-reasoning models allow you to tailor the "thinking" yourself. Just a feeling; not sure if this is accurate.

@thdxr What’s the problem with reasoning models? Haven’t they been responsible for most of the code quality improvements?

i still don't like reasoning models and it feels like that's the only way they're able to squeeze out gains not sure if base models are going to get much smarter - which is fine - i'm more interested in them getting cheaper and faster

@tyler_agg prompt

I just made a full walkthrough that makes it easy to write prompts for GPT and Claude that consistently get better results inside, you’ll get: – a simple structure that works across any use case – what context to include to get sharper answers – ways to make outputs sound personal and not robotic – easy techniques to avoid hallucinations and bad outputs comment “prompt” and I’ll DM you the full guide (must be following)

@tyler_agg track

this framework makes sure your AI is giving accurate answers by catching hallucinations I just made a full walkthrough on how to evaluate if your AI system is actually working... inside I show: – how I track AI output with a simple dashboard – the key signals I watch to catch mistakes early – how I test for accuracy with real-world questions – what to do when the AI gives a wrong answer – how to improve it over time using real feedback comment “track” and I’ll DM you the guide (must be following)

@jamesqquick MCP is just that, a protocol, a way of doing things. In this case, an API that has "self-discovery" endpoints with which you can know what functions the API has (tools, resources, etc) with a focus on LLM-applicable functionalities.

@elblancoaudaz There nothing inherent about MCP servers that make them well documented? With any API, it's depending on the person creating the documentation?

MCP is just like...an API right?

@thdxr Perhaps in a standard workflow, the default could be 1, like listing a new spec requirement - the interrupt option seems more like fixing or re-directing where the current task is going.

lot of requests for message queuing in opencode would you expect sending a message to a busy session to 1. wait until its done with current task then auto send next message 2. interrupt current task with new message

@aaditsh Who is this? Would be cool if you appended the original source of the video. Thanks!

@nilsxor @stevelauda_ open source?

I’m making a military radar for live tracking aircraft, just for fun. Inspired by @stevelauda_ design

@ImadTaieb @perplexity_ai API everything then, haha.

@elblancoaudaz @perplexity_ai Talking about myself, i use the desktop apps of these tools. The browser (chromium based ones) is not resources efficient, beside that from UX perspective it’s too limiting

Comet is here. A web browser built for today’s internet.