I think I'm the only one kind of stoked about this. My kiddos are going to LOVE making short films with their favorite Disney Princesses.

Yeah, but Disney will make you pay extra for it, that's for sure.

I used the Kernel Density Estimation (KDE) page/blog at my very first job. It was immensely useful and I've loved his work ever since.

Have you explored using Apple's javascript core engine at all? I know bun was built on it, but I don't know much else about it.

Not really. I've written a bunch of code to try maintain the limited support for it that already exists in GodotJS, but I've never really tried it. Main reason I haven't is I'm dependent on Web Worker(-like) APIs in GodotJS, and they're currently missing for JavaScript Core. But since I actually wrote some of those APIs, that's not really an excuse, I can port them easily enough.

So, yeah, I should really give it a shot. Thanks for the reminder.

I've been messing around with an Elixir + BEAM based agent framework. I think a mixture of BEAM + SQLite is about as good as you can get for agents right now.

You can safely swap out agents without redeploying the application, the concurrency is way below the scale BEAM was built for, and creating stateful or ephemeral agents is incredibly easy.

My plan is to set up a base agent in Python, Typescript, and Rust using MCP servers to allow users to write more complex agents in their preferred programming language too.

you should check out the Extism[0] project and the Elixir SDK[1]. This would allow you to write the core services, routing, message passing, etc in Elixir, and leverage all the BEAM/OTP have to offer, and then embed "agents" written in other languages which are small Wasm modules that act like in-process plugins.

[0]: https://github.com/extism/extism [1]: https://github.com/extism/elixir-sdk

That's a really interesting idea. My original thought was to use MCP as the way to define other agents, but I'll have to do some more research into extism!

Any reason for SQLite use, instead of the BEAMs built-in mnesia data store?

https://www.erlang.org/doc/apps/mnesia/mnesia.html

I'm still in the exploration/experimentation stage of the project, but I'm currently using a mixture of SQLite, PostgreSQL, S3, and DuckDB.

My original thought was to spin up SQLite databases as needed because they are super lightweight, well-tested, and supported by almost every programming language. If you want to set up an agent in another programming language via MCP, but you still want to be able to access the agent memory directly, you can use the same schema in a SQLite database.

I may end up using mnesia for more metadata or system-oriented data storage though. It's very well designed imo.

But one of the biggest reasons has just been the really nice integration with DuckDB. I can query all of the SQLite databases persisted in a directory and aggregate some metadata really easily.

It seems MUCH better at tool usage. Just had an example where I asked Sonnet 4 to split a PR I had after we had to revert an upstream commit.

I didn't want to lose the work I had done, and I knew it would be a pain to do it manually with git. The model did a fantastic job of iterating through the git commits and deciding what to put into each branch. It got everything right except for a single test that I was able to easily move to the correct branch myself.

When Grok 3 was released, it was genuinely one of the very best for coding. Now that we have Gemini 2.5 pro, o4-mini, and Claude 3.7 thinking, it's no longer the best for most coding. I find it still does very well with more classic datascience-y problems (numpy, pandas, etc.).

Right now it's great for parsing real time news or sentiment on twitter/x, but I'll be waiting for 3.5 before I setup the api.

See a example full in a few commands using uv think "wow I bet that Simon guy from twitter would love this" ... it's already him.

My favorite language is Python, but I wouldn't write a blog post about it because no one would care.

My favourite language is also python, and I would love to read your blog post on why your favourite language is python :-)

I had Grok summarize + evaluate the first chapter with thinking mode enabled. The output was actually pretty solid: https://pastebin.com/pLjHJF8E.

I wouldn't be surprised if someone figured out a solid mixture of models working as a writer (team of writers?) + editor(s) and managed to generate a full book from it.

Maybe some mixture of general outlining + maintaining a wiki with a basic writing and editing flow would be enough. I think you could probably find a way to maintain plot consistency, but I'm not so sure about maintaining writing style.

Would the kotlin equivalent just be: val dateRegex = Regex("""(\d{4})-(\d{2})-(\d{2})""") val (year, _, _) = dateRegex.matchEntire("2004-01-20")!!.destructured println("$year was a good year for PLs.") ?

The Scala snippet works for all inputs, it's something you would see on a real codebase. It also scopes the regex captures within the match expression. The code also scales better for more complex cases where a string is tested against multiple regexes, or against specific captured values.

I assume this Kotlin snippet would, at minimum, need to do null-checking on "matchEntire" before it finds its way to a production codebase.