fp.

Thoughts on Markdown

http://tantek.pbworks.com/w/page/59905776/Markdown

1•spacebuffer•50s ago•0 comments

Anthropic to limit Using third-party harnesses with Claude subscriptions

1•guiyuwei•3m ago•0 comments

Block's Dorsey Outlines AI-Powered Vision to Cut Middle Managers

https://www.bloomberg.com/news/articles/2026-03-31/block-s-dorsey-outlines-ai-powered-vision-to-c...

1•doctaj•4m ago•1 comments

Overview Effect

https://en.wikipedia.org/wiki/Overview_effect

1•andyjohnson0•4m ago•0 comments

Unofficial Telegram client 'Nekogram' turned out to be a spyware

https://thebadinteger.github.io/nekogram-phone-exfiltration/

2•mathfailure•7m ago•0 comments

Lightweight IDE to Pair with Claude Code?

1•zupancik•8m ago•1 comments

Writing an LLM from scratch, part 32h – Interventions: full fat float32

https://www.gilesthomas.com/2026/04/llm-from-scratch-32h-interventions-full-fat-float32

1•gpjt•8m ago•0 comments

1400% surge in GitHub commits explains service issues

https://twitter.com/kdaigle/status/2040164759836778878

2•tonymet•9m ago•0 comments

Y Combinator's CEO says he ships 37,000 lines of AI code per day

https://www.fastcompany.com/91520702/y-combinator-garry-tan-agentic-ai-social-media

2•jcbhmr•10m ago•0 comments

Imagination is more than sensory replay: study

https://news.feinberg.northwestern.edu/2026/03/31/imagination-is-more-than-sensory-replay/

2•hhs•11m ago•0 comments

Why Are BART and MUNI always broke(n)?

https://aakash.substack.com/p/why-are-bart-and-muni-always-broken

1•logicx24•13m ago•0 comments

Where do AI-built apps usually break when moving from prototype to production?

https://openbaton.com/

1•kellonedwards•13m ago•1 comments

Third-party Claude harnesses will now draw from extra usage

2•iBelieve•15m ago•0 comments

Resurrecting Supermaven: Trad coding is not dead. Just different

https://github.com/nhlmg93/supertab.nvim

1•nhelmig93•16m ago•1 comments

Show HN: SpeechSDK – free, open-source SDK that unifies all AI voice models

https://www.speechsdk.dev/

3•PiersonMarks•17m ago•0 comments

Compliance at scale and why TAM distracts with Christina Cacioppo of Vanta

https://cheekypint.substack.com/p/compliance-at-scale-and-why-tam-is

1•hhs•17m ago•0 comments

Trump announces 'fraud' crackdown in Democratic states as arrests begin in CA

https://www.theguardian.com/us-news/2026/apr/03/trump-vance-fraud-arrest-crackdown-california

4•mitchbob•18m ago•1 comments

Forest Service Will Close Research Stations That Study Wildfire Risk

https://www.nytimes.com/2026/04/03/climate/forest-service-research-stations.html

2•mitchbob•19m ago•0 comments

Anthropic no longer allowing Claude Code subscriptions to be used with OpenClaw

13•firloop•21m ago•4 comments

Researchers develop new way to detect breakthroughs in science: study

https://www.binghamton.edu/news/story/6153/eureka-scientists-develop-new-way-to-detect-breakthrou...

1•hhs•25m ago•0 comments

Tokenizer That Outperform Tiktoken with O200k_base

https://o200k-tokenizer-70fe25.gitlab.io/

1•nispin•26m ago•0 comments

Tinnitus: At a crossroad between phantom perception and sleep (2022)

https://academic.oup.com/braincomms/article/4/3/fcac089/6563428

3•gnabgib•34m ago•0 comments

AI startup envisions '100M new people' making videogames

https://www.pcgamer.com/software/ai/this-ai-startup-envisions-100-million-new-people-making-video...

3•solomonyardley•36m ago•2 comments

Meta Pauses Work with Mercor After Data Breach Puts AI Industry Secrets at Risk

https://www.wired.com/story/meta-pauses-work-with-mercor-after-data-breach-puts-ai-industry-secre...

5•srameshc•42m ago•1 comments

'Hacks' Star Hannah Einbinder Blasts AI Creators as 'Losers'

https://variety.com/2026/tv/news/hannah-einbinder-ai-creators-losers-1236706302/

2•bogzz•45m ago•1 comments

Show HN: Standalone TurboQuant KV Cache Inference

https://github.com/g023/turboquant

3•g023•45m ago•1 comments

Darkness can "travel" faster than light speed

https://www.nature.com/articles/s41586-026-10209-z

4•Jimmc414•46m ago•2 comments

Soros: The anatomy of an agentic geopolitical simulation engine

https://asksoros.com/insights/soros-anatomy-agentic-geopolitical-simulation

4•muggermuch•46m ago•4 comments

Fake Fans

https://www.wordsfromeliza.com/p/fake-fans

9•performative•50m ago•0 comments

The Subprime AI Crisis Is Here

https://www.wheresyoured.at/the-subprime-ai-crisis-is-here/

15•dmitrygr•51m ago•2 comments

Thoughts on Markdown

Anthropic to limit Using third-party harnesses with Claude subscriptions

Block's Dorsey Outlines AI-Powered Vision to Cut Middle Managers

Overview Effect

Unofficial Telegram client 'Nekogram' turned out to be a spyware

Lightweight IDE to Pair with Claude Code?

Writing an LLM from scratch, part 32h – Interventions: full fat float32

1400% surge in GitHub commits explains service issues

Y Combinator's CEO says he ships 37,000 lines of AI code per day

Imagination is more than sensory replay: study

Why Are BART and MUNI always broke(n)?

Where do AI-built apps usually break when moving from prototype to production?

Third-party Claude harnesses will now draw from extra usage

Resurrecting Supermaven: Trad coding is not dead. Just different

Show HN: SpeechSDK – free, open-source SDK that unifies all AI voice models

Compliance at scale and why TAM distracts with Christina Cacioppo of Vanta

Trump announces 'fraud' crackdown in Democratic states as arrests begin in CA

Forest Service Will Close Research Stations That Study Wildfire Risk

Anthropic no longer allowing Claude Code subscriptions to be used with OpenClaw

Researchers develop new way to detect breakthroughs in science: study

Tokenizer That Outperform Tiktoken with O200k_base

Tinnitus: At a crossroad between phantom perception and sleep (2022)

AI startup envisions '100M new people' making videogames

Meta Pauses Work with Mercor After Data Breach Puts AI Industry Secrets at Risk

'Hacks' Star Hannah Einbinder Blasts AI Creators as 'Losers'

Show HN: Standalone TurboQuant KV Cache Inference

Darkness can "travel" faster than light speed

Soros: The anatomy of an agentic geopolitical simulation engine

Fake Fans

The Subprime AI Crisis Is Here

Quirky table tennis equipment promotes "more creative" play at French school

Thoughts on Markdown

Anthropic to limit Using third-party harnesses with Claude subscriptions

Block's Dorsey Outlines AI-Powered Vision to Cut Middle Managers

Overview Effect

Unofficial Telegram client 'Nekogram' turned out to be a spyware

Lightweight IDE to Pair with Claude Code?

Writing an LLM from scratch, part 32h – Interventions: full fat float32

1400% surge in GitHub commits explains service issues

Y Combinator's CEO says he ships 37,000 lines of AI code per day

Imagination is more than sensory replay: study

Why Are BART and MUNI always broke(n)?

Where do AI-built apps usually break when moving from prototype to production?

Third-party Claude harnesses will now draw from extra usage

Resurrecting Supermaven: Trad coding is not dead. Just different

Show HN: SpeechSDK – free, open-source SDK that unifies all AI voice models

Compliance at scale and why TAM distracts with Christina Cacioppo of Vanta

Trump announces 'fraud' crackdown in Democratic states as arrests begin in CA

Forest Service Will Close Research Stations That Study Wildfire Risk

Anthropic no longer allowing Claude Code subscriptions to be used with OpenClaw

Researchers develop new way to detect breakthroughs in science: study

Tokenizer That Outperform Tiktoken with O200k_base

Tinnitus: At a crossroad between phantom perception and sleep (2022)

AI startup envisions '100M new people' making videogames

Meta Pauses Work with Mercor After Data Breach Puts AI Industry Secrets at Risk

'Hacks' Star Hannah Einbinder Blasts AI Creators as 'Losers'

Show HN: Standalone TurboQuant KV Cache Inference

Darkness can "travel" faster than light speed

Soros: The anatomy of an agentic geopolitical simulation engine

Fake Fans

The Subprime AI Crisis Is Here