Every analyzed article

Current Rothko (rothko.joonas.wtf)

Anecdotes of Wellington - The Wellington Society Of Madrid (wellsoc.org)

LFM2.5-8B-A1B: an Even Better on-Device Mixture-of-Experts (liquid.ai)

Bill C-22 Is a Mess of the Government’s Own Making (ethanplant.ca)

SQLite is All You Need for Durable Workflows (obeli.sk)

Robinhood now lets your AI agents trade stocks (techcrunch.com)

GitHub - facebookresearch/atlas-lean: ATLAS Autoformalized Textbook Library At Scale (github.com)

The Science of Weather and the Nature of Science (the-hinternet.com)

wterm | Terminal Emulator for the Web (wterm.dev)

Someone used my open source project to phish 14,000 people (andrej.sh)

Notes from the AI Now Summit by Mistral (koenvangilst.nl)

CAPTCHAs can still detect AI agents (research.roundtable.ai)

Someone used my open source project to phish 14,000 people (andrej.sh)

Step 3.7 Flash — A high-efficiency Flash model for Real-World (static.stepfun.com)

The Dead Economy Theory (owenmcgrann.com)

‘We cannot ban our way out of a youth mental health crisis’: social media bans for teenagers lack evidence and pose risks, scientists say - Science news (frontiersin.org)

GTA 6 Developers Announce Rockstar Games Union (rockstarintel.com)

The Secret Garden of Rock-Paper-Scissors (theshamblog.com)

Strengthening societal resilience with Rosalind Biodefense (openai.com)

It's hard to justify buying a Framework 12 (jeffgeerling.com)

bijou64 (inkandswitch.com)

It Will Never Be the Year of the Linux Desktop · unix.foo (unix.foo)

Expertise in the Age of AI (moderndescartes.com)

Free Furigana Converter: Kanji to Hiragana | EZFurigana (ezfurigana.com)

Headway Therapy Patients Forced to Scan Their Faces to Keep Getting Care (404media.co)

Poll: How often do you check "newest"?

Tulip mania: when a single flower was worth more than a house (dutchreview.com)

Please Use AI (shawnsmucker.substack.com)

GitHub - scanaislop/aislop: Catch the slop AI coding agents leave in your code. 40+ rules, 7 languages, sub-second, deterministic, no LLM. MIT. (github.com)

Even (very) noisy LLM evaluators are useful for improving AI agents · TensorZero (tensorzero.com)