Offshore
Photo
NecoKronos
Spot orderbook is starting to show some passive support, but the futures side is getting heavy with shorts piling up.
Interesting shift on the aggregate heatmap too: we're seeing liquidity being pulled from 65k and restacked lower at 62k.
I’m still eyeing that 65k zone for a potential rotation, but if the floor gives way, the next futures support is clearly defined lower.
#BTC
tweet
Spot orderbook is starting to show some passive support, but the futures side is getting heavy with shorts piling up.
Interesting shift on the aggregate heatmap too: we're seeing liquidity being pulled from 65k and restacked lower at 62k.
I’m still eyeing that 65k zone for a potential rotation, but if the floor gives way, the next futures support is clearly defined lower.
#BTC
tweet
Offshore
Photo
NecoKronos
Big day for Hong Kong crypto:
✅ Stablecoin licenses coming in March.
✅ Institutional perpetual contracts framework confirmed.
✅ $BTC & $ETH now eligible as collateral for broker financing.
The liquidity floodgates for institutional capital are officially opening. Watch the BTC and ETH dominance closely 👀
#BTC
tweet
Big day for Hong Kong crypto:
✅ Stablecoin licenses coming in March.
✅ Institutional perpetual contracts framework confirmed.
✅ $BTC & $ETH now eligible as collateral for broker financing.
The liquidity floodgates for institutional capital are officially opening. Watch the BTC and ETH dominance closely 👀
#BTC
tweet
Offshore
Photo
God of Prompt
RT @rryssf_: MIT researchers taught an LLM to write its own training data, finetune itself, and improve without human intervention
the paper is called SEAL (Self-Adapting Language Models) and the core idea is genuinely clever
but "GPT-6 might be alive" is not what this paper says. not even close.
here's what it actually does:
tweet
RT @rryssf_: MIT researchers taught an LLM to write its own training data, finetune itself, and improve without human intervention
the paper is called SEAL (Self-Adapting Language Models) and the core idea is genuinely clever
but "GPT-6 might be alive" is not what this paper says. not even close.
here's what it actually does:
tweet
Offshore
Photo
God of Prompt
RT @godofprompt: I've written 500 articles, 23 whitepapers, and 3 ebooks using Claude over 2 years, and these 10 prompts are the ONLY ones I actually use anymore because they handle 90% of professional writing better than any human editor I've worked with and cost me $0.02 per 1000 words: 👇 https://t.co/Yx6MCNdLbr
tweet
RT @godofprompt: I've written 500 articles, 23 whitepapers, and 3 ebooks using Claude over 2 years, and these 10 prompts are the ONLY ones I actually use anymore because they handle 90% of professional writing better than any human editor I've worked with and cost me $0.02 per 1000 words: 👇 https://t.co/Yx6MCNdLbr
tweet
Offshore
Photo
Javier Blas
French President Emmanuel Macron met this week with a group of seven newspapers for a joint interview.
He was asked about the Spanish blackout (and the role of the cross-border electricity connections between the Iberian peninsula and France).
He didn't mince his words: https://t.co/4si62pdfwx
tweet
French President Emmanuel Macron met this week with a group of seven newspapers for a joint interview.
He was asked about the Spanish blackout (and the role of the cross-border electricity connections between the Iberian peninsula and France).
He didn't mince his words: https://t.co/4si62pdfwx
tweet
Michael Fritzell (Asian Century Stocks)
RT @LowAlphaHighVol: @MikeFritzell WT: Tongshu Wang
LyGH: Grace Lu
Bin Yuan: Ping Zhou
FengHe: Matt Hu
Golden Pine: She Peng
Willing: Jacky Lin
Perseverance: Deng Xiaofeng
Trivest: Huimin Wu
Tairen: Larry Chen (in his heyday, was the OG)
3W: Weiwei Wu
tweet
RT @LowAlphaHighVol: @MikeFritzell WT: Tongshu Wang
LyGH: Grace Lu
Bin Yuan: Ping Zhou
FengHe: Matt Hu
Golden Pine: She Peng
Willing: Jacky Lin
Perseverance: Deng Xiaofeng
Trivest: Huimin Wu
Tairen: Larry Chen (in his heyday, was the OG)
3W: Weiwei Wu
tweet
God of Prompt
R E A D
T H I S
N O W
O R
Y O U
W I L L
R E G R E T
I T
L A T E R
tweet
R E A D
T H I S
N O W
O R
Y O U
W I L L
R E G R E T
I T
L A T E R
https://t.co/ivXRKXJvQg tweet
X (formerly Twitter)
Matt Shumer (@mattshumer_) on X
Something Big Is Happening
Jukan
Rapidus plans to begin mass-producing 2nm wafers at a rate of 6,000 wafers per month by the end of 2027, and ramp up to 25,000 wafers per month by the end of 2028.
According to the business plan Rapidus submitted to Japan's Ministry of Economy, Trade and Industry (METI), mass production of 2-nanometer chips is set to begin in the second half of fiscal year 2027.
The company plans to scale production to four times the 2027 level by 2028.
At the start of mass production in the second half of FY2027, Rapidus plans to produce 6,000 wafers per month, with the goal of expanding capacity to approximately 25,000 wafers per month within one year.
tweet
Rapidus plans to begin mass-producing 2nm wafers at a rate of 6,000 wafers per month by the end of 2027, and ramp up to 25,000 wafers per month by the end of 2028.
According to the business plan Rapidus submitted to Japan's Ministry of Economy, Trade and Industry (METI), mass production of 2-nanometer chips is set to begin in the second half of fiscal year 2027.
The company plans to scale production to four times the 2027 level by 2028.
At the start of mass production in the second half of FY2027, Rapidus plans to produce 6,000 wafers per month, with the goal of expanding capacity to approximately 25,000 wafers per month within one year.
tweet
God of Prompt
RT @godofprompt: The longer you wait to jump into AI, the bigger the gap between you and early adopters.
Not because AI gets harder. Because it gets easier.
More seamless. More plug-and-play. More black box.
Right now you can still see how it works under the hood. Context engineering. Open-source models. Building your own workflows in n8n.
Early internet users who configured dial-up and navigated BBS boards didn’t just understand networking better.
They became the founders, CTOs, and architects who built the companies everyone else ended up working for.
The people who understood HTTP, DNS, and FTP in 1995 weren’t just “tech savvy.”
They were the ones who saw ecommerce, SaaS, and cloud infrastructure coming before those words even existed.
The same thing is happening with AI right now.
The people tinkering with prompts, agents, and open-source models today aren’t wasting time on tools that’ll feel primitive in 5 years.
They’re building the intuition that lets them architect what comes next.
The DIY window doesn’t stay open forever.
tweet
RT @godofprompt: The longer you wait to jump into AI, the bigger the gap between you and early adopters.
Not because AI gets harder. Because it gets easier.
More seamless. More plug-and-play. More black box.
Right now you can still see how it works under the hood. Context engineering. Open-source models. Building your own workflows in n8n.
Early internet users who configured dial-up and navigated BBS boards didn’t just understand networking better.
They became the founders, CTOs, and architects who built the companies everyone else ended up working for.
The people who understood HTTP, DNS, and FTP in 1995 weren’t just “tech savvy.”
They were the ones who saw ecommerce, SaaS, and cloud infrastructure coming before those words even existed.
The same thing is happening with AI right now.
The people tinkering with prompts, agents, and open-source models today aren’t wasting time on tools that’ll feel primitive in 5 years.
They’re building the intuition that lets them architect what comes next.
The DIY window doesn’t stay open forever.
tweet
Offshore
Photo
God of Prompt
Prompt engineering is dead.
"Prompt chaining" is the new meta.
Break one complex prompt into 5 simple prompts that feed into each other.
I tested this for 30 days. Output quality jumped 67%.
Here's how to do it ↓ https://t.co/K3r7wrcJw7
tweet
Prompt engineering is dead.
"Prompt chaining" is the new meta.
Break one complex prompt into 5 simple prompts that feed into each other.
I tested this for 30 days. Output quality jumped 67%.
Here's how to do it ↓ https://t.co/K3r7wrcJw7
tweet