I recently found the Brooks Falls bear cam (Katmai, Alaska). It's become the calmest thing on my screen.

It's peak salmon run right now, so it's basically bears and their all-you-can-eat salmon feast.

But there's a bear that only fishes by belly-flopping at the salmon, and as far as I can tell has never once caught anything. I have no idea how it's still eating.

A few others just sit in the water doing nothing at all.

Here it is: https://www.youtube.com/watch?v=J7ZrIDvqlic

What's yours?

Hi HN, we built world-model-optimizer, an open source tool to continually improve models specialized to agents.

Agent traces you already capture are opportunities to get signal on how to make your model cheaper, faster, better.

We do this by continuously a) distilling relevant chain of thought from larger open source models into smaller ones, b) model routing to frontier + OS models, and c) token compaction to remove noise and save on tokens.

`wmo build` allows you to build a simulation to optimize against with your agent traces with your OpenRouter key

`wmo optimize` trains a router, compaction, and distills chain of thought from a larger model into your specialized model

`wmo serve` gives you an endpoint for your model

When you call your model, behind the scenes a router decides which tasks should go to the frontier versus your model. Tinker continually trains as new traces arrive.

We're also working on a hosted solution that does continual training + serving for you https://experientiallabs.ai

SpinWin is an open source menu bar app to visually rotate any individual window. You can rotate / spin as many as you want one by one.

I was scrolling reels when one came on where the content creator suggested "turning the phone upside down" for a portion. I was sitting at my mac, and decided to see if I could find something that did this. To my surprise, nothing exists, so I decided to build it.

Turns out macOS doesn't provide any public API to do anything like this, so I wound up using the Accessibility API to move the target window offscreen, where it keeps rendering. Then we capture it's contents with ScreenCaptureKit, and display that on a transparent, borderless overlay window which has a CALayer to rotate it.

It's written in Swift, open source, MIT license, and signed/notarized.

Have fun - would love to hear your comments.

i am an incoming msc student thinking about a research direction in reinforcement learning. which rl subfields do you think have the most potential right now? i am particularly interested in emboddied ai and bcis.