Gabe Pereyra ✓

@gabepereyra · joined 21 Feb 2022

building @harvey with my bud @winstonweinberg

11 887Followers

237Following

414Posts total

1.8MViews on collected posts

Against accounts of the same size

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

Median views59 318this account2 896median for 10K–100K

Reach, %499.01%this account9.51%median for 10K–100K

Engagement, %0.41%this account1.56%median for 10K–100K

Metric	This account	Median for 10K–100K	Ratio
Median views per post	59 318	2 896	20.5×
Reach (views ÷ followers)	5.0× audience	9.51%	52.5×
Engagement rate	0.41%	1.56%	0.27×

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

178.4K11 Aug

58K

60.6K14 Aug

7.7K

646.9K17 Aug

14.2K

815.9K20 Aug

31.4K

Last 8 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.

Engagement rate per post

0.24%11 Aug

0.58%

0.15%14 Aug

1.13%

0.11%17 Aug

0.81%

0.11%20 Aug

0.77%

Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 1.56%.

What the audience does

Likes53.3%2 335 in total

Reposts6.8%297 in total

Replies2.7%119 in total

Quotes3.4%148 in total

Bookmarks33.8%1 481 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

Update on @harvey’s model training effort. We post-trained a model we are calling Tenet: - Achieves SOTA on LAB - Generalizes 3rd party legal benchmarks - Uses sub-agents for domain specific capabilities Tenet uses Kimi K3 as base and was post-trained in collaboration 31.4K views · 193 likes · 32 reposts · 10 replies 20 Aug 2026 https://t.co/1c2Dn0vuoX 815.9K views · 635 likes · 119 reposts · 37 replies 20 Aug 2026 Awesome research by @engramlab using the synthetic law firm we built. They trained a 27B Qwen model on synthetic client matters and used the learned knowledge to do online search more efficiently. The trained model outperforms frontier models at 10x lower cost per query. The 14.2K views · 98 likes · 11 reposts · 5 replies 17 Aug 2026 Today we're publishing our first research blog, Understanding a Law Firm through Study. We're sharing a glimpse of a future where agents are trained with native memory: https://t.co/XpKq5l3fHO 646.9K views · 601 likes · 65 reposts · 30 replies 17 Aug 2026 In July, 20% of our inference spend was review tables. The most expensive review table queries cost $20k. Today we’re excited to share work we did with @appliedcompute that will reduce this cost by 50% by post training a review table model. Our review table product allows 7.7K views · 74 likes · 8 reposts · 4 replies 14 Aug 2026 https://t.co/FLzObAJnod 60.6K views · 71 likes · 8 reposts · 1 replies 14 Aug 2026 Had so much fun giving this talk at @sequoia about @harvey’s moneyball approach to building a research lab. The biggest mistake I made in the early days of Harvey was trying to play the Yankees baseball style of frontier intelligence. I found out the hard way that we were the 58K views · 286 likes · 30 reposts · 17 replies 11 Aug 2026 Want world class research capabilities, but don’t have the resources of a big lab? At our recent Sovereign AI event, @gabepereyra shared @harvey ’s “moneyball” approach. Here’s the playbook: 00:00 Introduction 00:37 Building a research lab on a budget 02:28 Legal Agent Bench, 178.4K views · 377 likes · 24 reposts · 15 replies 11 Aug 2026