Interesting...
@OpenAI released support for skills over MCP from the SEP-2640 draft
they like to ship early (like when they released chatgpt apps before MCP Apps)
this is validation on how useful this feature is going to be!
@ThugDebugger@RenderATL in no particular order...
Sarah Drasner, Kelsey Hightower, Scott Hanselman, Erick Wendel, Paul Frazee, Charlie Gerard, April Gittens, Anthony Mays, Danny Thompson, Francesco Ciulla, mf me!
I get giddy when AI unlocks entirely NEW ways of working, not just faster versions of old ones.
Ex:
In 2024, for 3 straight months, I had a 10min voice conversation with AI in the mornings while I got ready
Ex:
To build out my second brain beyond meeting transcripts and emails and slack and drive and whatnot these days, I dictate to AI for a few minutes every day to get out the “not codified” parts of my work and mind
Ex:
I managed my finances but not at this level - I built out a revenue engineering workflow to run through every dollar of my business every week, break down the source and velocity, AND run pre-mortems and predictions based on social/email/meeting signals
I’m not trying to copy human workflows. If that was all I did, that would be a failure.
I've had people complain that I no longer talk about developing software the way I used to.
Do you actually want me to make videos about building accessible forms in React?
I'm sorry, but that's such a boring and useless thing to talk about now.
One of the easiest ways to stand out when writing or creating videos or posting:
DONT. USE. AI.
Suddenly you sound distinctively different from 80%+ ppl who do and output the same verbose + nicely formatted stuff that feels all the same (+ most ppl stop paying attention midway)
Today, key parts of the EU AI Act go into effect for companies whose AI systems produce output used in the EU.
If your company is building AI agents, you need to understand these rules.
I broke down what teams need to know and what they should do next.
https://t.co/dpUlSMrDoM
We're starting to leave the territory where you'd test an LLM by e.g. "create an svg of pelican on a bicycle". As one idea to generalize it, I was interested what Opus 5 would do if I gave it the first paragraph of the Lord of the Rings, a 1M token budget (~$10) and asked for three js render of it. Opus went off for ~2 hours and wrote 5500 lines of code that (procedurally) rendered the story. It's kind of janky but fun. But it's a bit mindboggling that the LLM has to place and orchestrate various polygon assets in (x,y,z) coordinates and write code that animates it all, and that it even does anything at all.
I also like this kind of examples because no one in their right mind would ever spend the time to write something this custom but LLMs have all the stamina and patience in the world, so it's an example where we go from "no one would ever do this" to "sure, why not, it's ~free". There might be a lot more. But I'm excited about creating hyper custom worlds that you can imagine dropping players into, e.g. here to participate in the LoTR story as a spectator NPC, or one of the characters, or etc. Something like an ephemeral GTA of X on demand.
Last thought is that the domain of worlds/games exposes a weakness in LLMs: they can't easily audit their work because they aren't able to efficiently and natively perceive videos or play games within them. Here, Opus 5 had to very slowly and painstakingly take screenshots at different points, and it messed up a few times and created a bunch of jank. An example of raw capability (multimodal, gameplay) that I think is still quite lacking.
You know that idea you've been putting off? You should build it! I'll even give you an excuse.
I'm hosting the Ready, Spec, Ship Hackathon alongside @johncrickett and @gregorojstersek where your challenge is to create something with Kiro that solves a genuine problem.
You and Kiro can jam it out yourselves, or you can grab a few more humans to join in on the fun. Have more than one idea? Submit them all!
I'm giving you a heads up... you have a week to think of your idea. Then the competition runs from August 8 - 23.
@_chenglou I was part of that team. Basically ChatGPT one year before it came out. Called LMChat and then another codename.
Google was too nervous to release it and DeepMind was blocked from shipping products that could disrupt Google.
I think about this a lot.
@precognita@gdb that's interesting. that's literally the EA's job though. In the past, my EAs have gotten upset with I try to book stuff for myself. In fact, I usually mess up something they were balancing/scheduling on my behalf by trying to do it myself