prompt_hater

Full Stack AI Engineer | Alum @IIITDelhi

[dax @thdxr]

it's not done if it's not implemented
it's not done if the implementation is ugly
it's not done if it's not documented
it's not done if users can't discover it
it's not done if you can't market it

[Dwarkesh Patel @dwarkesh_sp]

Re the Fable ML sandbagging, the model's AI research capabilities were probably at least partly trained on Anthropic employees diffing atop proprietary algos and infra.

So the IP leak is somewhat like a researcher who knows Anthropic's stack getting poached to another lab.

Anthropic's recent "When AI builds itself" post talks about a next-step eval. Where they snapshot a research session at the moment a human researcher made a suboptimal next-step choice, show a model only the transcript up to that point and ask what it would do next, then have a hindsight-equipped LLM judge decide whether the model's suggestion or the human's actual choice was better.

This eval seems like a very good RL target for AI R&D - one among many that could be used to have AIs emulate Anthropic researchers and their research products.

I'm just speculating. But if this was a motivation, then Anthropic should have figured out a better way to protect IP than sandbagging without telling the user they're sandbagging, which is very hostile and untrustworthy behavior.

[Lucas Beyer (bl16) @giffmana]

Actually it's fine guys! I figured out a way, see below.

Claude Fable 5 is a great model afterall, and I also finally appreciate the difference between CLAUDE.md and AGENTS.md.

It's all good.

[matt @MattVMacfarlane]

Was using Fable 5 to write my world model training code.

Anthropic flagged it as frontier AI research.

The steering vector kicked in and it started implementing JEPA 🤨

[Matt Pocock @mattpocockuk]

Everyone's banging on about loops

When they should be thinking about queues

[AVB @neural_avb]

My RLM library sitting at 400 stars!

[Siddharth @Pseudo_Sid26]

College students can spend 20 lakhs on Btech but cant spend even 2000 rupees per month on claude code or codex to work on projects.

Let your parents know about the AI HYPE and fcking tell them its part of the curriculum and fcking get it and learn to use it, you would not want to get left behind.

The industry wants people who can ship fast, not who can write good code in 2 months.

[Noam Brown @polynoamial]

https://t.co/oWqzT12RtZ

[Peter Steinberger 🦞 @steipete]

@matijaoe my slop is better than your slop.

[Patrick Jiang @patpcj]

Introducing Harness-1, a 20B search agent trained with a state-externalizing harness.

> frontier-level long-horizon search, rivaling Opus-4.6 and outperforming GPT-5.4

> Context-1-level cost and latency

> externalizes candidates, evidence, verification, and search history

> open-source

[Ben Lang @benln]

Pulled the fastest-growing startups based on X follower growth over the past 90 days:

[Rahul @sairahul1]

This is the best site on the internet to learn harness engineering.

Free. Completely.

Most AI engineers have never heard the term.

https://t.co/bwDbTTYsjM

Bookmark this site.

Then read this setup ↓

[Auriel @aurielws]

Modern RL hiring increasingly expects full-stack understanding. If you are an algorithm researcher, people will still ask infrastructure questions. The reverse is also true.
— Yep and for practical reasons as many times the harness is why things won’t converge ~~~

[rody @0x_rody]

Anthropic engineer James Brady:

"Every agent in production lies. We measured it. The good ones lie less, the great ones catch the lie before the user does."

In 29 minutes, he walks through the verification stack he built and the patterns the Claude Code team adopted to keep agents honest at scale.

Watch the full talk, then save the config below👇

[GREG ISENBERG @gregisenberg]

I was once pitching in a board room at a top 3 VC firm for a $15M Series A.

12 people in the meeting. One of the GPs fully fell asleep. Out cold for 30+ minutes. Nobody acknowledged it. Everyone just kept going.

I kept presenting my Series A slides to an unconscious man in a Herman Miller chair and somehow that was considered normal. That's venture capital.

You might fly across the country to perform for people who may or may not be conscious.

It's a dance.

And sometimes you lead and sometimes you follow and sometimes your partner is unconscious.

If you're raising right now, just know: every founder has a story like this. The process is weird. The power dynamic is weird. You're not crazy for thinking it's weird.

No one talks about it because they want to continue raising. But I'm happy to stick my neck out there.

It is weird.

[Karri Saarinen @karrisaarinen]

Common @linear workflow we have internally:
from @SlackHQ message to merged code in minutes.

User asked about MCP team docs support, Linear agent checked the code to verify if it's true then started coding session to add it. Code was then reviewed, improve and merged through Diffs.

[dax @thdxr]

my worst VC story:

[unnamed] partner stopped me mid pitch. this was pre-covid so these were all in person

he walked up to me and whispered in my ear "damn ur a hot piece of ass"

he smacked my butt and said he wanted my whole seed round

i was offended and left his bedroom immediately

[Ben Lang @benln]

Skip LinkedIn. Resources to find breakout startups hiring before everyone else:

• Ramp’s monthly vendor reports
• Harmonic’s quarterly Hot 25
• a16z Build newsletter
• Founders You Should Know
• Next Play newsletter
• YC startup directory
• Early Days Substack

[AyalKarmi @CoolestAK87]

Agent-ready ≠ scrapable.

Agent-ready = your site exposes typed tools (search, checkout, inventory)

Install the extension, pick one, and watch an agent actually use it: https://t.co/qrK6oT78BL

[David Ondrej @DavidOndrej1]

"Coding agents are eating software" @skirano
(ex-Anthropic)

Pietro Schirano reveals the Codex setup that 10x'd his speed, text commands, multi-agent spawning, agent-first design

He even built a game for a Flipper Zero with one prompt

Here is the full episode: