frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Open in hackernews

AI Loop Engineering in 2026: How to Build a Gauntlet Loop

https://www.thepromptindex.com/ai-loop-engineering-gauntlet-loop-guide.html
2•westurner•58m ago

Comments

westurner•48m ago
TIL the "Gauntlet loop" prompt engineering method came from development of the "Claude of Duty" three.js game;

R3F (ReactThreeFiber), Drei, and NextJS (with OpenNext) save time on generated interactive 3d visuals.

I've coded a few apps with ThreeJS recently as well.

First, I upgraded the ThreeJS particles WebGL example by hand for a number of code test iterations. In the months since, I've been working on superfluid quantum gravity; now with normed tensor gaussian splatters from 4th order tensors to 3rd for 3d and then an affine transformation to 2D.

I have an idea for a Snake-like game to demo the gaussian splatters and amplituhedron-enhanced twistor geometry in a game rendering engine ("splatsnake") that could be built with gauntlet loop ai loops instead of test coverage and loops like I've been running.

westurner•37m ago
(2 weeks ago) "Guantlet Loop" by Matt Shumer https://somethingbig.ai/gauntlet-loop .. https://news.ycombinator.com/item?id=49080957
westurner•24m ago
How to prevent wasteful AI loops without a human in the loop to intercede in death spirals that burn token budgets? Not unique to this loop pattern.

There was this from DeepMind a few weeks ago about online rewriting of agent prompt materials:

"Continual Harness: Online Adaptation for Self-Improving Foundation Agents" (2026) https://arxiv.org/abs/2605.09998 .. https://news.ycombinator.com/item?id=48126112

Costed WASM opcodes would probably help prevent waste by enabling optimization of agent loops.

Ask HN: How much Adderall do you use?

1•roschdal•1m ago•0 comments

Understanding an AI agent's memory through Harry Potter

https://squidler.io/blog/agent-memory-harry-potter
1•tidbeck•1m ago•0 comments

My home internet died for half a day. So did every agent I had

https://bostrat.ai/blog/the-outage
1•globocodes•2m ago•0 comments

Fedora CoreOS to Enable Systemd-Oomd and ZRAM Swap by Default

https://www.phoronix.com/news/Fedora-CoreOS-OOMD-zRAM-Swap
1•Bender•3m ago•0 comments

Ask HN: When should you open-source your solution and why?

1•Yahyaaa•3m ago•0 comments

TDD inside the agent loop – theater or actual value?

https://martinfowler.com/articles/exploring-gen-ai/tdd-in-the-agent-loop.html
1•groomlake•4m ago•0 comments

Show HN: HyperSAE – Hyperbolic Sparse Autoencoders for LLM Interpretability

https://github.com/vishal-dehurdle/hypersae
1•visha1v•4m ago•0 comments

LTX 2.5 a stronger foundation for what's being built

https://ltx.io/model/ltx-2-5
1•lastdong•5m ago•0 comments

Ask HN: How do you keep 54 LLM workflows on the right models?

1•ttruett•6m ago•0 comments

OpenAI's head of ethics leaves startup less than one year after joining

https://www.ft.com/content/e49dfb75-f841-4466-a577-f7aaff8779a0
4•aanet•8m ago•1 comments

A Preliminary Study on Simultaneous Coscheduling for Discrete GPU vs. Fused GPU

https://arxiv.org/abs/2608.09647
1•matt_d•11m ago•0 comments

Show HN: SmokeOps – describe a flow, get checks you rerun without AI

https://github.com/gate3/SmokeOps
1•crackedev•11m ago•0 comments

Imprint – Fine-tune MoE LLMs bigger than your RAM

https://github.com/sigma0101111/imprint
1•pyeAI•12m ago•0 comments

Sam's News – Free AI News Aggregator

https://samnews.news/
1•srschreiber•13m ago•1 comments

We Used to Get Jobs

https://ironicsans.ghost.io/how-we-used-to-get-jobs/
2•speckx•14m ago•0 comments

Nvidia building 1T-parameter Nemotron 4 to rival open AI models

https://www.reuters.com/business/nvidia-is-developing-nemotron-4-open-source-models-information-r...
2•giuliomagnifico•14m ago•0 comments

Cynthion: Open-source USB test instrument

https://greatscottgadgets.com/cynthion/
1•kaycebasques•14m ago•0 comments

Elon Musk ran a decoy super PAC to compare Trump with RGB

https://www.google.com/search?q=rgb+pac+elon+musk
4•mandeepj•16m ago•0 comments

The worst part of using multiple LLM providers is not the API differences

https://github.com/maximhq/bifrost
1•Swapnoneel•16m ago•0 comments

A bug becoming a feature (1972 edition)

https://unsung.aresluna.org/a-bug-becoming-a-feature-1972-edition/
1•leephillips•16m ago•0 comments

Ethical Cold Outreach

https://blog.val.town/ethical-cold-outreach
1•stevekrouse•17m ago•0 comments

Report from the Security Mines

https://blog.val.town/security
1•stevekrouse•17m ago•0 comments

Docfooding: Eating Our Own Documentation

https://blog.val.town/docfooding
1•stevekrouse•17m ago•0 comments

Hack's Law

https://en.wikipedia.org/wiki/Hack%27s_law
1•brandonb•17m ago•0 comments

Would Appreciate Any Feedback

https://deskfellow.app/
1•nunep•19m ago•1 comments

ChatGPT Desktop App for Linux

https://twitter.com/OpenAI/status/2087231350134980830
3•tosh•20m ago•0 comments

Why Are Rivers So Mathematical?

https://www.quantamagazine.org/why-are-rivers-so-mathematical-20260810/
2•brandonb•21m ago•0 comments

(2,1)-C1P is NP-complete

https://zenodo.org/records/21871667
1•birriel•21m ago•0 comments

Inaccessible .bnnsir files on macOS Sequoia

https://lapcatsoftware.com/articles/2025/4/3.html
2•xoa•21m ago•1 comments

How to Appreciate

https://how-to-appreciate.pages.dev/
1•srid•21m ago•0 comments