@BrendanFoody profile picture
Brendan (can/do) @BrendanFoody

ceo @mercor_ai | organizing human intelligence

Joined March 2020
  • Posts 631
  • Following 505
  • Followers 23,158
27 Photos and videos
Pinned Post
Brendan (can/do)
 @BrendanFoody
Jun 6
AI progress requires (1) compute, (2) algorithms, and (3) data. - The leading compute company is worth $5 trillion. - The leading model company is worth $1 trillion. - @mercor_ai is the leading data company and is currently valued orders of magnitude lower. There's an opportunity in how the market is mispricing the value of data. Data is the oil of the AI revolution. It is the primary way that models and enterprises build competitive advantages.
23
23
381
86,128
Nvidia should have bought TSMC in the summer of 2024 for ~20% of its market cap
8
100
16,382
I keep hearing people argue that continual learning on enterprise usage data will solve the data bottleneck for foundation models. No Fortune 500 will allow this. Their enterprise data is their competitive advantage. Authorizing a foundation model to train on it is like giving a competitor access to your Slack workspace and GitHub repository.
33
5
227
45,202
do you think George Washington would support working from home?
19
1
163
30,798
10% regression for Fable 5 is crazy
Mercor
 @mercor_ai
Jul 3
Fable 5 is back and we’ve got results for the re-released version on APEX-SWE. While it did not perform as well as its earlier version from June, the model still significantly outperforms Opus 4.8. Fable 5 (June): 65.5% Pass@1 Fable 5 (July): 54.8% Pass@1 Opus 4.8: 45.3% Pass@1 This re-release scored about 10 points below the original Fable 5, however it still beat Opus 4.8 by more than 9 points.
11
3
228
58,498
Mercor now pays out over $4M / day to experts on our platform, with an average pay rate of over $100 / hour. Despite today's jobs report missing estimates, training agents is becoming one of the fastest-growing job categories in the world. Organizing human intelligence is the largest bottleneck to AI progress.
15
24
375
53,737
Loom has regressed so much since the acquisition by Atlassian. It's the fastest product degradation I've seen after an acquisition, rivaling what Salesforce did to Slack.
68
6
541
89,701
Brendan (can/do) reposted
Mercor
 @mercor_ai
Jul 1
GLM 5.2 just became the first open-source model to lead a category on APEX-SWE. It scored a 55.3% Pass@1 on Integration, the top score we've recorded for any model, open or closed source. On the overall leaderboard, GLM 5.2 scored 37.3% Pass@1, ranking 6th place. That makes it the best open-source model we've tested on APEX-SWE to date. Right behind it is Kimi K2.7 from Moonshot AI, now the second-best open-source model on the APEX-SWE leaderboard. Congrats to @Zai_org and @Kimi_Moonshot on two strong model releases.
10
35
355
22,823
Brendan (can/do) reposted
Mercor
 @mercor_ai
Jun 30
Claude Sonnet 5 ranks #3 on APEX-SWE with 43.7% Pass@1. We tested @claudeai Sonnet 5 on APEX-SWE which tests AI models on real software engineering work. It’s another step forward in Anthropic's lineup, placing just behind Opus 4.8 and well ahead of the rest of the field. Fable 5 remains their top performing model. Congrats to @AnthropicAI on the release!
Claude
 @claudeai
Jun 30
Introducing Claude Sonnet 5, our most agentic Sonnet yet. It makes plans, uses tools like browsers and terminals, and runs autonomously at a level that just a few months ago required larger and more expensive models.
0:11
14
8
115
36,898
Brendan (can/do) reposted
Liz Wessel
 @lizwessel
Jun 30
What are the most overlooked companies today that are most likely to produce the next "PayPal Mafia" (ie alum who go on to create iconic bizs)? Not the obvious answers (ramp, mercor, etc)... I'm curious for the non-obvious picks. 👀
56
2
94
56,083
Out of all of the founders I've met who @benchmark has invested in over the past 5 years, their founder NPS is among the highest in Silicon Valley
Marc Andreessen 🇺🇸
 @pmarca
Jun 29
Cc @travisk Adam Neumann @jack @naval @emilmichael @bhorowitz
8
3
161
80,225
the Mercor Mafia is so cracked
13
5
150
38,683
Brendan (can/do) reposted
Aravind Srinivas
 @AravSrinivas
Jun 27
Every enterprise will have its own model-harness-sandbox-eval flywheel with token value per watt optimization. This is the future. Simple reason: tacit knowledge about the domain and customers and their workflows that the company uniquely understands and has built trust around.
115
153
1,787
223,646
99% of human knowledge is in people’s heads. It’s usually not written down. It’s certainly not published on the open internet. To achieve superintelligence, we need to teach it everything we know.
14
7
150
15,498
Brendan (can/do) reposted
TBPN
 @tbpn
Jun 26
a16z's @bhorowitz says the California wealth tax is the "best strategy" to dismantle Silicon Valley that he's ever seen. "It's been so hard to break the Silicon Valley network effect, but this is the best strategy I've seen." "Norway has an unrealized capital gains tax. Norway's got a lot of extremely smart people, great entrepreneurs, but they all left." "When you talk to entrepreneurs in Norway, they're like, 'I literally can't pay the tax because the company got marked up, and I own a lot of it, and I can't get that money out.'" "'It's a private company. I'm stuck. So, I have to leave the country.'" "There are basically no tech entrepreneurs in Norway now." From his appearance on the show in January.
2:44
52
152
1,187
162,867
Office liveliness at 8pm on a Friday correlates so much with startups success. Most people who enjoy their jobs enjoy spending more time in the office. People who don’t leave at 5
20
9
261
52,422
This is the best way to end America's entrepreneurial culture. Most founders I know endure 100 hours of work/week because they are obsessed with their mission and the impact they can have on the world. They make < $100K / year, fly economy, and live modest lives. The average American doesn't want their life. Just because their company's valuation is growing doesn't mean they are "wealthy". When you impose a > $100M tax on illiquid stock, they will declare bankruptcy or flee the country.
Gavin Newsom
 @GavinNewsom
Jun 26
It's time for a national billionaires tax and a new social contract. 10% of Americans own two-thirds of the wealth. Wages have stagnated. The cost of living has skyrocketed.  The system is fundamentally broken. The federal tax code, a corporate code, and an inheritance code were written for a different set of Americans.  It’s time for an economic reset. substack.com/@gavinnewsom
2:17
106
103
1,246
292,405
Brendan (can/do) reposted
Hardik Bhatnagar
 @hrdkbhatnagar
Jun 25
New #1 on PostTrainBench: GLM 5.2 (Max reasoning) hits 34.29%, narrowly beating Opus 4.8 Max (34.08%) What makes GLM 5.2 interesting: zero failed runs across 84 runs (vs ~10% failure rate for Opus agents). The most reliable agent we've seen Leaderboard: posttrainbench.com
14
60
498
103,587
Brendan (can/do) reposted
Harry Stebbings
 @HarryStebbings
Jun 22
Why Remote Work is White Collar Fraud. "I have a three-year-old and a five-year-old. The idea that I could do any work at my house is like a total fantasy. The kids come home at 3pm, your work day needs to keep going. I'm highly against it." @typesfast
0:30
1,069
45
1,227
5,396,874
We forecast Mercor's long-term growth based on the % of global salaries allocated to creating data/training agents. As more and more work shifts towards training agents, this will become increasingly clear.
Brendan (can/do)
 @BrendanFoody
Jun 24
Training agents will be the largest job category in 5 years. At Mercor, many of our teams already spend majority of their time training agents to complete repetitive tasks. Instead of people responding to repetitive customer support tickets or watching interviews, they train agents to automate those. This takes many forms. Sometimes training just requires updating context. Other times, it requires building environments to update the weights. Meta is another good example. Tens of thousands of their employees now create RL tasks, but the interface will need to change. Majority of my time is training colleagues on our team, giving them feedback to improve their work. In the future we’ll train agents in a dynamic, collaborative way, similar to how we train our colleagues.
4
2
133
22,750
Load more