iiinova — where your digital self keeps evolving

Reposted by @wsmoak.bsky.social

Katie Mack @astrokatie.com · 01/10/2026

I went to Geneva a few weeks ago and spoke on a panel about the nature of time for a Geneva Watch Days event. While I was there, on a lark, I entered a competition to most precisely calibrate a Swiss watch. And today I found out that I won! 😃 www.instagram.com/reel/Dd6hUTB...

Me peering through a little magnifier at a tiny watch mechanism connected to some monitoring equipment, while making adjustments with a small screwdriver

39710735598

wsmoak.bsky.social @wsmoak.bsky.social · 21/09/2026

Yes, Jev is "just" machine learning in the same way LLMs are "just" math. Before: license a rules engine and spend months configuring it, or train your own classifier. Now it's a hosted typed API call that's easy to use. And it's been fun to see what everyone came up with over the weekend.

011

Reposted by @wsmoak.bsky.social

Ant Stanley @iamstan.dev · 20/09/2026

Built a thing... well my agents built a thing. decide CLI to use with Jev... use it yourself, or hand it to your agent to use with it's bash tool. Still experimental, and WIP. Today its a wrapper around the API with some auth handling. github.com/antstanley/d...

github.com

GitHub - antstanley/decide

Contribute to antstanley/decide development by creating an account on GitHub.

231

wsmoak.bsky.social @wsmoak.bsky.social · 19/09/2026

Jev for sales call transcripts -- decide whether or not the prospect explicitly asked for each one of our features, whether they asked for something we do not have, whether the sales rep said no when we actually DO have it -- in a single, fast, API call. github.com/wsmoak/jev-s...

github.com

GitHub - wsmoak/jev-sales-calls: Example: TypeSafe AI's Jev model reading sales call transcripts against a product capability library

Example: TypeSafe AI's Jev model reading sales call transcripts against a product capability library - wsmoak/jev-sales-calls

000

wsmoak.bsky.social @wsmoak.bsky.social · 26/06/2026

See also: “Why this matters”

000

Reposted by @wsmoak.bsky.social

Craig Pittman @craigtimes.bsky.social · 05/06/2026

#Florida county clerk announces that a historic book listing cattle brands has been preserved and, even better, digitized so the public can review it online. www.theledger.com/story/news/h...

theledger.com

Polk County book of historic cattle brands preserved and digitized

Polk County has digitized a book of historic cattle brands from the 1860s to the 1940s, preserving local ranching history.

1172

wsmoak.bsky.social @wsmoak.bsky.social · 12/04/2026

Upgraded all the local AI things and trying out Gemma 4 -- good results so far in OpenClaw (web search, Fitbit stats) and Open WebUI including with images flic.kr/p/2s7acHm

flic.kr

gemma4-26b-image-20260412

Trying out gemma4:26b in Ollama via Open WebUI

000

Reposted by @wsmoak.bsky.social

David Ho @davidho.bsky.social · 26/03/2026

I don't think people fully appreciate how apocalyptic things are for US science. I haven't had any new funding since 2024, but I'm still ok since typical grants are for three years. This means next year I will be completely out of funding and will have to fire everyone in the lab. It's not great.

4153272362

wsmoak.bsky.social @wsmoak.bsky.social · 22/03/2026

Claude is figuring out how to integrate LangChain's OpenSWE with DevPod as the backend for sandboxes. It is slowly making progress... There is _zero_ chance any of this would be happening if I had to read all the code and docs and figure it all out by myself.

Claude session showing it using subagents to build, push, deploy and watch the task status before it proceeds with commenting on a GitHub issue to trigger a webhook.

000

wsmoak.bsky.social @wsmoak.bsky.social · 21/02/2026

Adventures in local LLMs with a Titan RTX card - cat pictures included wsmoak.net/2026/02/21/l...

000

Reposted by @wsmoak.bsky.social

Santiago Greco @fsg.bsky.social · 15/02/2026

Having a linguistics background I think there's a huge misconception when it comes to prompting LLM's, the vast majority thinks that coding is "hard" compared with speaking a natural language. Well it's actually the contrary from a utilitarian POV less verbs at your disposal => easier to be specific

2526

wsmoak.bsky.social @wsmoak.bsky.social · 30/12/2025

> ⁠The difference between “AI helped me ship” and “AI gave me a mess to untangle” almost never comes down to just model selection. It comes down to context engineering, prompt specificity, workflow structure, and guardrails. -Addy Osmani addyo.substack.com/p/how-good-i...

addyo.substack.com

How Good Is AI at Coding React (Really)?

A data-driven look at what AI can and can’t do for React developers - and what you can do about it

000

wsmoak.bsky.social @wsmoak.bsky.social · 09/11/2025

Open Source in 2025: clone a git repo, discover it's in a language I've never used, install the language, build the project, try the examples, realize it's missing a feature I want, have Claude Code implement the missing feature. So now I guess I get to learn Rust and see if the code is correct!

010

Reposted by @wsmoak.bsky.social

Tim 🎃wens @attimowens.bsky.social · 08/06/2025

lol

A Toyota Highlander, with a stick figure decal underneath the marque, depicting four headless figures and a fifth, head firmly attached, holding a sword, with the words "THERE CAN BE ONLY ONE."

01017

Reposted by @wsmoak.bsky.social

Randall Munroe @xkcd.com · 31/05/2025

Internet Roadtrip is my favorite thing on the internet right now—it’s basically Twitch Plays Google Street View. They started in Boston, spent several weeks on a quest to cross the Canadian border, and are now exploring New Brunswick. neal.fun/internet-roa...

neal.fun

Internet Roadtrip

Let's take a streetview roadtrip

744236739

Reposted by @wsmoak.bsky.social

Reuven M. Lerner @lernerpython.com · 28/05/2025

Does #Python pass arguments by value? Or by reference? I'm asked this question a lot, and the answer isn't a simple "yes" or "no." (Which is even more confusing!) In my latest video, I explain how you should think about it, and why neither term is quite right: youtu.be/Ah1J85nCdNo

youtu.be

Understanding call-by-value and call-by-reference in Python

YouTube video by Python and Pandas with Reuven Lerner

051

wsmoak.bsky.social @wsmoak.bsky.social · 17/05/2025

Are you using OTel Ruby SDK logging in production, particularly for a Rails application? It's still marked Development, but... it seems to work and the community is responsive. Is it safe enough to give it a try at this point, or would you have reservations? #otel #observability #ruby #rails

000

Reposted by @wsmoak.bsky.social

Ryan Michalesko @ryanmichalesko.bsky.social · 07/04/2025

There are so many cranes on the Platte that it can be a fight for space—as one flock takes flight, another fills in their spot. On our last morning on the river, researchers conducting aerial surveys estimated an all-time record of 736,000 Sandhill Cranes on the 80-mile stretch of the Platte River.

0696

wsmoak.bsky.social @wsmoak.bsky.social · 01/12/2024

Anthropic's Model Context Protocol + #Claude Desktop is another level. In less than an hour I had Claude write an #MCP server for retrieving data from the YNAB (You Need a Budget) API. Now I can ask it about account balances and spending by category. It just works.

030

wsmoak.bsky.social @wsmoak.bsky.social · 11/11/2023

Thanks to a note on Substack with some invite codes, I have arrived. There appear to be an acceptable number of cat pictures in the Discover feed, and also @neilhimself.neilgaiman.com is here. So far so good.

tabby cat loafing in a sunny upstairs window

050