The Few Bets That Matter
$NVDA didn’t put PCIe v6 connectors on Rubin GPUs just to look pretty.

They’re showing optimization potential. This is the next massive vertical. Optimizing hardware for energy consumption and maximum compute.

$ALAB offers this. Through PCIe v6.

Yet, the market sells the stock as it assumes optimization will be fully integrated into $NVDA GPUs, tomorrow.

Maybe someday. But not tomorrow, we're years away from this, if it ever happens.

Vera Rubin will be on volume production end of 2026 and run through 2027. So yeah, not tomorrow. Until then, optimization will come from hird-party hardware. From $ALAB. And for years.

After that, they'll offer optical and other optimizing hardware, so $ALAB has a very long and profitable path in front of himself.

The market is selling the stock on bad assumptions. Understand the overreaction. Take advantage.
https://t.co/mm3A4EAyJG
tweet

God of Prompt
hey @grok, analyze all upcoming AI breakthroughs in 2026 using first principles thinking and research and compile into a concise bullet point list comprising of 7 key elements
tweet

God of Prompt
RT @godofprompt: nano banana prompt:

{
"subject": {
"description": "A young woman sitting on yoga mat, wiping sweat with towel, holding water bottle",
"mirror_rules": "N/A - direct gym photo",
"age": "late 20s",
"expression": "accomplished, slight breathlessness, confident smile",
"hair": {
"color": "blonde with highlights",
"style": "high ponytail, slightly messy with flyaways from workout"
},
"clothing": {
"top": {
"type": "sports bra",
"color": "dusty rose pink",
"details": "medium support, strappy back detail, moisture visible from sweat"
},
"bottom": {
"type": "high-waisted leggings",
"color": "black with mesh panels",
"details": "ankle length, mesh cutouts on calves, compression fit"
}
},
"face": {
"preserve_original": true,
"makeup": "minimal, dewy from workout, natural flushed cheeks, no eye makeup"
}
},
"accessories": {
"headwear": {
"type": "none",
"details": "hair pulled back in scrunchie"
},
"jewelry": {
"earrings": "small diamond studs",
"necklace": "none",
"wrist": "rose gold fitness tracker, black hair ties on wrist",
"rings": "none"
},
"device": {
"type": "smartphone",
"details": "propped against dumbbell, recording workout selfie"
},
"prop": {
"type": "insulated water bottle",
"details": "matte black 32oz bottle with motivational quote sticker, condensation visible"
}
},
"photography": {
"camera_style": "gym selfie aesthetic, smartphone front camera",
"angle": "slightly above eye level, sitting position",
"shot_type": "full upper body and crossed legs, centered composition",
"aspect_ratio": "9:16 vertical",
"texture": "crisp detail, bright gym lighting, energetic feel"
},
"background": {
"setting": "modern gym studio",
"wall_color": "light gray with motivational mural",
"elements": [
"purple yoga mat laid out",
"set of dumbbells scattered nearby",
"white towel draped over shoulder",
"blurred gym equipment in background",
"large mirror reflecting back wall",
"resistance bands coiled on floor"
],
"atmosphere": "energetic, accomplished, health-focused",
"lighting": "bright overhead LED gym lighting, even coverage"
}
}
tweet

Illiquid
$sndk $mu - looks like I can recycle my “if you missed storage and memory you should buy this” posts. https://t.co/NSE97z0JjU
tweet

Quiver Quantitative
BREAKING: We just caught another crazy trade.

Someone on Polymarket just bet $40,000 that Trump will acquire Greenland.

They will win $300K if they are correct.

Another potential insider or a degenerate gambler? https://t.co/zLfxmaK3Qz
tweet

God of Prompt
the best content creators aren't writers.

they're librarians.

they know where to find information.
they know how to organize it.
they know how to resurface it at the right time.

creation is curation with context.
tweet

God of Prompt
RT @godofprompt: R.I.P few-shot prompting.

Meta AI researchers discovered a technique that makes LLMs 94% more accurate without any examples.

It's called "Chain-of-Verification" (CoVe) and it completely destroys everything we thought we knew about prompting.

Here's the breakthrough (and why this changes everything): 👇
tweet

Wasteland Capital
$MU now passed the 300%+ return mark. Meanwhile $AEO just hit new 4-year highs.

$LYFT outperforming +55-60% vs $UBER and $DASH (both of those are negative).

Average return now +139%.

Compounding is beautiful. https://t.co/ZerFN5dOf0

Apart from my earnings reviews, I’ve only tweeted five new cases on here over the last year.

They ended up the #1 Mag7 YTD $GOOG, the #1 Semi YTD $MU & #1 China LargeCap YTD $BABA. Plus 2 🚀 smallcaps, $AEO & $LYFT

Average return +111% currently.

Less is more, as they say. https://t.co/LGr5bglq7t - Wasteland Capital
tweet

Illiquid
Wrong kind of storage to be checking out today. https://t.co/mTqMenqxLE
tweet

Brady Long
RT @0xgaut: Mom: "how's the job search going?"

you: "Claude, build me a prediction market arbitrage bot" https://t.co/eFnRqOu05Y
tweet

God of Prompt
RT @godofprompt: 🚨 DeepMind discovered that neural networks can train for thousands of epochs without learning anything.

Then suddenly, in a single epoch, they generalize perfectly.

This phenomenon is called "Grokking".

It went from a weird training glitch to a core theory of how models actually learn.

Here’s what changed (and why this matters now):
tweet