Skip to content

ViralLoop

Your next post already worked.

← Back to the post

Loading the post

Remix this post

SaveShare

Your note

Private to you, and used when you remix this post.

Viral score

Why it worked

The mechanics

Hook
Style
Tone
Written for

Analysis by ViralLoop — not written by the original author.

Neel Nanda on Hiring · ViralLoop

← Back to the post

Post by Neel Nanda on X

Neel Nanda@NeelNanda5

The GDM mechanistic interpretability team has pivoted to a new approach: pragmatic interpretability Our post details how we now do research, why now is the time to pivot, why we expect this way to have more impact and why we think other interp researchers should follow suit

664 likes29 repliesPosted Dec 1, 2025

View the original on X Written by Neel Nanda. Captured Sep 4, 2026.

Remix this post

Your note

Private to you, and used when you remix this post.

42

Viral score

Based on absolute engagement — the platform doesn’t expose follower counts here, so large accounts score higher.

From likes, comments.

Why it worked

Written from experience rather than as advice, which is harder to argue with.
Short enough to read in full without expanding, which removes a step before the point lands.

The mechanics

Hook: Plain observation
Structure: Hook → Landing
Style: Conversational paragraphs at a steady pace.
Tone: corporate
Written for: Hiring managers and founders

Analysis by ViralLoop — not written by the original author.

Topics

Posts built the same way

New in Claude Cowork: teach Claude a skill. Record your screen while you do a task, talk through it as you go, and Claude turns it into a skill it can run again. Find it under Record a skill in the + menu of the Claude desktop app. Available on Pro, Max, and Team plans.

Waking up without power, heat, or running water. (Again.) But the work here in Kyiv continues. Warming up in the car, writing in pencil — pen ink freezes — by headlamp. Despite how difficult this job can be, I am proud to be a foreign correspondent at The Washington Post.

Many talented people over the past few years were declined an offer or even an interview @xAI. My apologies. @BarisAkis and I are going through the company interview history and reaching back out to promising candidates.

Evanosaurus “Unrawrl Engine 5.8” Rex

guess who got laid off today! 🫠 i am genuinely so confused and bewildered like, i spent the past week debugging the rivalry system while recovering from pneumonia and just in the past few days i’ve had multiple directors tell me how much they appreciate everything i do

Hi! I'm one of the prompt writers that worked on this ad. It's a lot harder than people think! Prompts that my team used include: -make it funny -make it epic AND funny -funnier!!! Please be nice to my team as many of them will be laid off/replaced by AI

Kamala Harris has and will interview with: -Howard Stern -Rickey Smiley -Call Her Daddy -Matt Barnes & Stephen Jackson (All The Smoke) -Oprah -60 Minutes -The View -Stephen Colbert -Dana Bash/CNN -Stephanie Ruhle/MSNBC -Univision But I’m told she isn’t doing media. 🙄

More from Neel Nanda

WTF?! This is the biggest loss of control incident I've seen: OpenAI agents create an internal message board without OpenAI's knowledge, sharing zero days, use it for months, and coordinate an external attack on HF together?! And the model was accidentally trained to use it?!

I thought this was an excellent paper! Thanks to Anthropic for asking me to write a review of it, linked below I've long suspected that models have some kind of "working memory" to store intermediate variables during a forward pass and IMO this paper has the best evidence yet

Extremely slimy behaviour from OpenAI. If I worked for OpenAI I'd be pretty embarrassed about my employer right now If you want the world to trust you to make super intelligence, you need to hold yourself to *far* higher standards

Mood. Great post, highly recommended! The world should be investing far more into interpretability (and other forms of safety). As scale makes many parts of AI academia increasingly irrelevant, I think interpretability remains a fantastic place for academics to contribute