Michael Fritzell (Asian Century Stocks)
RT @JoeValue: NOW FREE - My highest conviction idea 🇯🇵
Tobila Systems $4441.T $4441.JP
Smoak Capital (@dsmoak98) just published an excellent summary (link below).
Daniel at Smoak Capital has an exceptional 37.4% CAGR since 2018!
Smoak Capital owns 7.3% of Tobila.
Disclaimer: Personal opinions only, not investment advice.
Links to Smoak Capital Letter and my own writeup below:
tweet
RT @JoeValue: NOW FREE - My highest conviction idea 🇯🇵
Tobila Systems $4441.T $4441.JP
Smoak Capital (@dsmoak98) just published an excellent summary (link below).
Daniel at Smoak Capital has an exceptional 37.4% CAGR since 2018!
Smoak Capital owns 7.3% of Tobila.
Disclaimer: Personal opinions only, not investment advice.
Links to Smoak Capital Letter and my own writeup below:
tweet
Offshore
Photo
Fiscal.ai
Booking Holdings v. Airbnb
Total Bookings:
$BKNG: $186B, +16.3% YoY
$ABNB: $91B, +15.7% YoY https://t.co/u92vEzzHSC
tweet
Booking Holdings v. Airbnb
Total Bookings:
$BKNG: $186B, +16.3% YoY
$ABNB: $91B, +15.7% YoY https://t.co/u92vEzzHSC
tweet
Offshore
Photo
Fiscal.ai
DoorDash just reported its largest jump in orders ever.
Total Orders: 903M, +32% YoY
$DASH https://t.co/VQ4KE87XET
tweet
DoorDash just reported its largest jump in orders ever.
Total Orders: 903M, +32% YoY
$DASH https://t.co/VQ4KE87XET
tweet
Offshore
Photo
Fiscal.ai
Carvana just reported its lowest gross profit per unit in 8 quarters.
$CVNA: -23.3% AH https://t.co/oRq8t8TGNO
tweet
Carvana just reported its lowest gross profit per unit in 8 quarters.
$CVNA: -23.3% AH https://t.co/oRq8t8TGNO
tweet
Michael Fritzell (Asian Century Stocks)
RT @brentmuio: TIL Mixue Ice Cream & Tea https://t.co/8s3DQmETv4 is the largest restaurant chain globally by store count. 45k stores vs 42k for $MCD and 40k for $SBUX
tweet
RT @brentmuio: TIL Mixue Ice Cream & Tea https://t.co/8s3DQmETv4 is the largest restaurant chain globally by store count. 45k stores vs 42k for $MCD and 40k for $SBUX
tweet
Offshore
Photo
Clark Square Capital
Love this case study h/t @dirtcheapstocks
tweet
Love this case study h/t @dirtcheapstocks
My favorite Buffett investment: Delta Duck Club.
- $2.8mm market cap (4x earnings).
- 13% dividend yield.
- Oilfield remained active for 50+ years. https://t.co/hNiLZV2aBS tweet
Offshore
Photo
Fiscal.ai
Customers are spending more on Figma every year... a lot more.
The design and prototyping software just reported a net dollar retention rate of 136%.
$FIG https://t.co/Mw6euDlQyN
tweet
Customers are spending more on Figma every year... a lot more.
The design and prototyping software just reported a net dollar retention rate of 136%.
$FIG https://t.co/Mw6euDlQyN
tweet
Michael Fritzell (Asian Century Stocks)
RT @Basalt_Capital: I recently invested in Espec (https://t.co/8Glmwuvv0Q). Espec is the #1 producer of environmental test chambers with a global market share of 30%. These chambers assess the impact of temperature and humidity fluctuations on electronic components and systems throughout R&D. Management prioritizes higher-margin laboratory testing services over lumpy equipment sales and focuses on rapidly evolving AI semiconductors, autonomous driving, and satellite communications markets. Orders received in the target markets increased 83% YoY in the first half of 2025. Last November, Espec announced a JPY 3.5 billion stock buyback, equivalent to 4% of the total shares. I see 60-70% valuation upside. I think investors underestimate the improvements in Espec’s business model, growth opportunities, and capital allocation.
tweet
RT @Basalt_Capital: I recently invested in Espec (https://t.co/8Glmwuvv0Q). Espec is the #1 producer of environmental test chambers with a global market share of 30%. These chambers assess the impact of temperature and humidity fluctuations on electronic components and systems throughout R&D. Management prioritizes higher-margin laboratory testing services over lumpy equipment sales and focuses on rapidly evolving AI semiconductors, autonomous driving, and satellite communications markets. Orders received in the target markets increased 83% YoY in the first half of 2025. Last November, Espec announced a JPY 3.5 billion stock buyback, equivalent to 4% of the total shares. I see 60-70% valuation upside. I think investors underestimate the improvements in Espec’s business model, growth opportunities, and capital allocation.
tweet
Offshore
Photo
God of Prompt
RT @godofprompt: 🚨 Holy shit… Stanford just published a paper that questions whether we even need humans to study humans.
The title sounds like a joke:
“This human study did not involve human subjects.”
But it’s dead serious.
The researchers are asking a controversial question:
Can LLM simulations count as behavioral evidence?
Here’s the core idea.
Instead of recruiting thousands of participants, running surveys, and waiting weeks for results, they simulate people using large language models.
Not generic prompts.
But structured simulations where the model is assigned demographic traits, preferences, beliefs, and contextual constraints.
Then they test whether the simulated responses statistically match real-world human data.
And disturbingly… they often do.
Across multiple behavioral tasks, the LLM-generated “participants” reproduced known human patterns:
• Established psychological biases
• Preference distributions
• Decision-making trends
• Even demographic splits
Not perfectly. Not universally.
But far closer than most people would expect.
The key contribution of the paper isn’t “LLMs are human.”
It’s validation.
They systematically compare simulated outputs to ground-truth human datasets and evaluate alignment using statistical benchmarks.
When the distributions match, the simulation isn’t just storytelling.
It becomes empirical evidence.
That’s the uncomfortable shift.
If a sufficiently constrained LLM simulation reproduces real behavioral patterns, does it become a legitimate experimental proxy?
Because if the answer is yes, this changes everything:
• Behavioral economics
• Political science
• Market research
• Policy testing
• UX experimentation
You could prototype social interventions before deploying them in the real world.
You could stress-test messaging strategies across simulated demographics.
You could explore rare edge-case populations without recruitment bottlenecks.
But here’s where Stanford is careful.
The models don’t “understand” humans.
They reflect training data patterns.
They can amplify biases.
They can collapse under distribution shift.
And they can simulate plausibility without causality.
So the paper doesn’t claim replacement.
It argues for calibration.
LLM simulations can be useful behavioral instruments if validated against real data and bounded within known limits.
That’s the distinction.
Not synthetic humans.
Synthetic behavioral priors.
The wild part?
This paper forces academia to confront something bigger:
If large models encode large-scale behavioral regularities from the internet, they become compressed maps of human tendencies.
Not minds.
Maps.
And maps can be useful.
We’re moving from “AI as text generator” to “AI as behavioral simulator.”
The ethics, methodology, and epistemology implications are massive.
Because once simulation becomes statistically reliable, the bottleneck in social science shifts from data collection to model alignment.
And that might be the real revolution hidden in this paper.
tweet
RT @godofprompt: 🚨 Holy shit… Stanford just published a paper that questions whether we even need humans to study humans.
The title sounds like a joke:
“This human study did not involve human subjects.”
But it’s dead serious.
The researchers are asking a controversial question:
Can LLM simulations count as behavioral evidence?
Here’s the core idea.
Instead of recruiting thousands of participants, running surveys, and waiting weeks for results, they simulate people using large language models.
Not generic prompts.
But structured simulations where the model is assigned demographic traits, preferences, beliefs, and contextual constraints.
Then they test whether the simulated responses statistically match real-world human data.
And disturbingly… they often do.
Across multiple behavioral tasks, the LLM-generated “participants” reproduced known human patterns:
• Established psychological biases
• Preference distributions
• Decision-making trends
• Even demographic splits
Not perfectly. Not universally.
But far closer than most people would expect.
The key contribution of the paper isn’t “LLMs are human.”
It’s validation.
They systematically compare simulated outputs to ground-truth human datasets and evaluate alignment using statistical benchmarks.
When the distributions match, the simulation isn’t just storytelling.
It becomes empirical evidence.
That’s the uncomfortable shift.
If a sufficiently constrained LLM simulation reproduces real behavioral patterns, does it become a legitimate experimental proxy?
Because if the answer is yes, this changes everything:
• Behavioral economics
• Political science
• Market research
• Policy testing
• UX experimentation
You could prototype social interventions before deploying them in the real world.
You could stress-test messaging strategies across simulated demographics.
You could explore rare edge-case populations without recruitment bottlenecks.
But here’s where Stanford is careful.
The models don’t “understand” humans.
They reflect training data patterns.
They can amplify biases.
They can collapse under distribution shift.
And they can simulate plausibility without causality.
So the paper doesn’t claim replacement.
It argues for calibration.
LLM simulations can be useful behavioral instruments if validated against real data and bounded within known limits.
That’s the distinction.
Not synthetic humans.
Synthetic behavioral priors.
The wild part?
This paper forces academia to confront something bigger:
If large models encode large-scale behavioral regularities from the internet, they become compressed maps of human tendencies.
Not minds.
Maps.
And maps can be useful.
We’re moving from “AI as text generator” to “AI as behavioral simulator.”
The ethics, methodology, and epistemology implications are massive.
Because once simulation becomes statistically reliable, the bottleneck in social science shifts from data collection to model alignment.
And that might be the real revolution hidden in this paper.
tweet
Offshore
Photo
Fiscal.ai
Remitly Global has grown its active customers ~10X since 2019.
2019: 948k
2020: 1.9M
2021: 2.8M
2022: 4.2M
2023: 5.9M
2024: 7.8M
2025: 9.3M
That's a 46% CAGR over 6 years.
$RELY https://t.co/VvbQLgoaHA
tweet
Remitly Global has grown its active customers ~10X since 2019.
2019: 948k
2020: 1.9M
2021: 2.8M
2022: 4.2M
2023: 5.9M
2024: 7.8M
2025: 9.3M
That's a 46% CAGR over 6 years.
$RELY https://t.co/VvbQLgoaHA
tweet
Offshore
Video
God of Prompt
RT @godofprompt: Every client call I used to walk into blind.
No idea what their business does, what's broken, or what they actually need.
Now an AI agent briefs me before I even say hello.
New YT video dropped 👇
https://t.co/DHliL0hjLJ
tweet
RT @godofprompt: Every client call I used to walk into blind.
No idea what their business does, what's broken, or what they actually need.
Now an AI agent briefs me before I even say hello.
New YT video dropped 👇
https://t.co/DHliL0hjLJ
tweet