Hacker News
X / Twitter
@FardeemM @dexhorthy https://t.co/4fO7qbvAJC
Just finished creating a comprehensive library of algorithms used in pen #plotting! Every single one is interactive and generative. Enjoy and share!
published on @drawingbots website:
https://t.co/Us9d4u3Iwc
#svg #plottertwitter https://t.co/NumsvoMaJK
Hot take 🔥🔥: an enterprise context layer on top of existing items (semantic models, data, pipelines, emails, PDFs, metadata, etc.) that automatically improves with usage and, most importantly, is autonomous.
If this works, it could be the next big thing in analytics.
we’re really seeing the clawlake takeoff happening in real time:
- Cursor Origin
- Amp Orbs
- Pierre Computer
- Cloudflare Artifacts
any others i’m missing?
It's literally one prompt:
Get Azure Kinect depth working on my Apple Silicon Mac by running Microsoft's arm64 Linux depth engine in a QEMU/HVF VM with both the depth and color cameras passed through as root (the SDK opens the color MCU unconditionally), streaming depth+IR back over localhost TCP
Would cost me a lot more to OSS and wouldn't probably fit your case.
https://t.co/wWY9TICI8D
Every illustration in this dataset now has an instance mask. Think of it as 19th-century clip art: over a million public-domain images, cut out clean.
Pair it with the SigLIP2 embeddings (also in the dataset) and you (or your agent) can find an engraving for almost any topic and grab the cutout. No image generation required!
@kentdebruin I collected 70 different moats, and they're all still relevant
https://t.co/AnpR9P3bV6
I find the "inception" pattern to be very useful in many agentic use cases. You can force the model to take an action when it thinks for too long by injecting a thought after a specified reasoning budget.
Helps dealing with underspecified tasks which make the model reason for way too long.