tweetindex
EN

Walden ✓

@walden_yan · joined 12 Nov 2020

I sometimes code @cognition

14 276Followers
726Following
989Posts total
1.3MViews on collected posts

Latest posts

Fable cheaper than Opus -- what a crazy upset Cost improvements across Devin today 6.2K views · 65 likes · 1 reposts · 4 replies Open on X →
Introducing Fable 5.1 in Devin Fable-level intelligence is now 54% cheaper – making it even cheaper than Opus – due to a change in caching. Devin’s Fusion harness is now even smarter and cheaper than before, matching Fable 5.1 on FrontierCode at 47% lower cost. Here's how: http 206K views · 1.1K likes · 77 reposts · 68 replies Open on X →
Minor detail: I spent an embarrassing amount of time trying different harnesses for intelligent range detection and line folding Users don’t even notice it because it’s designed to be hidden. Until they go back to GitHub https://t.co/jwRNkJCR2J 258.2K views · 255 likes · 8 reposts · 4 replies Open on X →
smart diffs on @DevinAI is very very good, im surprised more agent harness companies have not done this as well. 18.9K views · 35 likes · 1 reposts · 0 replies Open on X →
Cool $1M initiative to make AI mathematics progress Excited to be supporting from @cognition 16.6K views · 93 likes · 6 reposts · 1 replies Open on X →
A few @Caltech friends and I are hosting the first math hackathon with frontier labs/startups. We're giving $1M+ of compute to 100 teams to solve the hardest open problems -- competitors range from IMO golds to frontier/neolab folks to math PhDs/profs. DM me if your company 139.1K views · 550 likes · 60 reposts · 32 replies Open on X →
Come work with us! 8.8K views · 101 likes · 0 reposts · 3 replies Open on X →
@walden_yan You should never share context with agents and agents should never communicate. You should issue commands to temporary subagents that contain reference docs and files, and that is also what the subagents should return. I've been building this since December. 2.8K views · 12 likes · 1 reposts · 2 replies Open on X →
@walden_yan While I appreciate how deep you went into all of this, there is one important thing to note: There is severe degradation when main agent passes on instructions to a subagent. But not because of an incorrectly set up system - I would argue most SOTA models will do f 2K views · 12 likes · 0 reposts · 1 replies Open on X →
@walden_yan it is a bad idea to make smaller model ask a smarter model. you will get: wrong trajectories, wasteful tokens spend of the bigger model, wrong questions, etc 926 views · 3 likes · 0 reposts · 0 replies Open on X →
https://t.co/iY0G5ieQhJ 644.9K views · 1.6K likes · 190 reposts · 28 replies Open on X →

Against accounts of the same size

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

Median views18 933this account924median for 10K–100K
Reach, %132.62%this account3.62%median for 10K–100K
Engagement, %0.61%this account1.52%median for 10K–100K
MetricThis accountMedian for 10K–100KRatio
Median views per post18 93392420.5×
Reach (views ÷ followers)132.62%3.62%36.6×
Engagement rate0.61%1.52%0.40×

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

644.9K22 Apr
926
2K
2.8K23 Apr
8.8K28 Aug
139.1K30 Aug
16.6K
18.9K31 Aug
258.2K
206K1 Sep
6.2K

Last 11 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.

Engagement rate per post

0.29%22 Apr
0.32%
0.65%
0.54%23 Apr
1.18%28 Aug
0.48%30 Aug
0.61%
0.20%31 Aug
0.11%
0.64%1 Sep
1.14%

Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 1.52%.

What the audience does

Likes43.8%3 861 in total
Reposts3.9%344 in total
Replies1.6%143 in total
Quotes1.5%133 in total
Bookmarks49.2%4 337 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.

Similar accounts

Frequently asked questions about @walden_yan

How many followers does @walden_yan have?

14 276 followers.

How many posts has @walden_yan published?

989 posts.

How do posts by @walden_yan perform?

Collected posts: 11. Views: 1 304 522. Likes: 3 861.