Aakanksha Chowdhery ✓

@achowdhery · Stanford, CA · joined 13 Mar 2011

@Stanford // Previously @GoogleDeepMind :: PaLM, Gemini, @reflection_ai // @MSFTResearch, @Princeton // views my own and subject to change

14 968Followers

7 242Following

564Posts total

116.6KViews on collected posts

Against accounts of the same size

2 posts from the last 90 days, next to the 10K–100K follower range. shown widely, but few of those viewers react.

Median views22 887this account980median for 10K–100K

Reach, %152.91%this account3.27%median for 10K–100K

Engagement, %0.96%this account2.01%median for 10K–100K

Metric	This account	Median for 10K–100K	Ratio
Median views per post	22 887	980	23.4×
Reach (views ÷ followers)	152.91%	3.27%	46.8×
Engagement rate	0.96%	2.01%	0.48×

Others in this range → Compare with another account → How these benchmarks are built →

Growth & engagement

How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.

Views per post

27.4K2 Dec

8.5K

35K16 Mar

40K5 Aug

5.8K

Last 5 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.

Engagement rate per post

0.52%2 Dec

0.73%

0.53%16 Mar

1.12%5 Aug

0.81%

Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 2.01%.

What the audience does

Likes62.4%769 in total

Reposts6.2%76 in total

Replies2.5%31 in total

Quotes0.5%6 in total

Bookmarks28.4%350 in total

Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.

The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.

Latest posts

Interesting analysis on compute supply/demand for AI inference. The world has ~19 million H100 equivalent AI chips. Even if 500 million knowledge workers, each with an AI that runs a modest 50-agent swarm, this needs about 830 million chips. Roughly 44 times every AI chip that 5.8K views · 36 likes · 5 reposts · 6 replies 05 Aug 2026 · Open on X →

I and Azalia Mirhoseini built and taught the course on Self Improving AI Agents (CS329A) twice at Stanford now. Very excited to bring the course to the broader community online. YouTube: https://t.co/mt1fMi7zZu Research is moving very fast in this field - here we have focused 40K views · 383 likes · 43 reposts · 18 replies 05 Aug 2026 · Open on X →

Building open intelligence matters because it democratizes the model weights for all and they can be inspected as well as customized independently. Today Reflection is proud to share that its partnership with Shinsegae Group to build a 250-megawatt AI factory for South Korea; 35K views · 160 likes · 16 reposts · 6 replies 16 Mar 2026 · Open on X →

Worth checking out the panel on Agentic AI/RL as well at 4pm! https://t.co/dCn5xQaDRY We talk about Agentic development at the frontier! 8.5K views · 56 likes · 5 reposts · 1 replies 02 Dec 2025 · Open on X →

Looking forward to being at #NeurIPS2025 this week from later tomorrow! Many of us from @reflection_ai will be there and would love to chat with people interested in joining us! 27.4K views · 134 likes · 7 reposts · 0 replies 02 Dec 2025 · Open on X →

#GoogleIO2022 showcases reasoning and multilingual capabilities of PaLM model. https://t.co/VV67gMNfBN Blog: https://t.co/662mRkcupI arXiv: https://t.co/GqJhSp6DC6 https://t.co/j5j9ydiAJx 0 views · 761 likes · 131 reposts · 19 replies 11 May 2022 · Open on X →