SGANT

[SGANT @sgantavande]

“Two Little Hearts” — 40-Second Cinematic Lovebirds Video

[SGANT @sgantavande]

10 AI tricks that can immediately make you a better AI user

Most people know how to ask AI questions.
Far fewer know how to make AI think with them.
Here are 10 techniques worth learning: 👇

1/10
Ask AI to challenge your assumptions.
Don't ask:
“Is my idea good?”
Ask:
“What would make this idea fail?”
That changes the conversation completely.

2/10
Ask for multiple solutions.
Instead of:
“Give me the best strategy.”
Try:
“Give me 5 different strategies and explain when each would work best.”

3/10
Make AI show its assumptions.
“List the assumptions you're making before giving me your recommendation.”
This can expose hidden weaknesses.

4/10
Give AI a role.
“Act as a skeptical investor.”
“Act as a lawyer reviewing this contract.”
“Act as a hiring manager.”
The perspective can dramatically change the output.

5/10
Give AI a scoring system.
“Score each option from 1–10 for cost, risk, difficulty and potential return.”
Now you're getting structured analysis instead of a wall of text.

6/10
Ask AI to critique its own answer.
“What's the weakest part of your previous answer?”
Then:
“What evidence would change your conclusion?”

7/10
Don't ask AI to do everything at once.
Research → outline → critique → draft → edit.
Breaking complicated tasks into stages usually produces better results.

8/10
Give it your actual context.
AI doesn't know what matters to you unless you tell it.
Budget.
Timeline.
Experience.
Goals.
Constraints.
Context = better decisions.

9/10
Use AI as a debate partner.
Tell it:
“Take the opposite position and argue against me.”
You may discover problems you hadn't considered.

10/10
And finally:
Verify important information.
AI can be incredibly useful.
But confidence ≠ accuracy.
The best AI users don't blindly trust AI.
They know when to question it.

Follow for practical AI knowledge without the hype.

[SGANT @sgantavande]

AI doesn't replace critical thinking.
It increases the value of it.

Why?

Because AI makes generating information incredibly cheap.
The scarce skill becomes:
Knowing what information is actually worth believing.

[SGANT @sgantavande]

A simple AI rule:
The more important the decision, the more you should challenge the AI.

Low stakes: “What's a good dinner recipe?”

High stakes: “Should I invest $100,000?”

The second question deserves verification, competing viewpoints and real sources.

[SGANT @sgantavande]

Want better AI answers?

Don't just say:
“Write me a business plan.”

Instead:
“You are a skeptical investor. Analyze this business idea. Identify the 5 biggest risks, assumptions that may be wrong, and reasons you would reject the investment.”

Give AI a role + objective + constraints + evaluation criteria.

[SGANT @sgantavande]

One of the most underrated AI skills:
Ask AI to disagree with you.
Try:
“Assume my conclusion is wrong. What are the strongest arguments against it?”
You’ll often learn more from the answer than from asking AI to validate your idea.

AI can make you faster.
But speed isn't the same as productivity.

If AI helps you produce 10 bad ideas instead of 1 good one, you've gained almost nothing.

The real skill is knowing what deserves to be done.

[SGANT @sgantavande]

🧵 5 things you should NEVER blindly trust AI to do:

Make important financial decisions
Give medical diagnoses
Interpret laws without verification

Provide current information without checking sources
Tell you what you want to hear
AI is powerful.
That doesn't make it infallible.

[SGANT @sgantavande]

1. AI isn't just about asking better questions.
The real advantage comes from knowing how to: → Give context
→ Set constraints
→ Challenge the answer
→ Ask for alternatives
→ Verify important claims
Prompting is only the beginning.

2. Most people use AI like Google.
They ask:

“What's the best investment?”

Power users ask:
“Compare 5 approaches, identify the assumptions behind each, challenge your own recommendation, then tell me what information would change your conclusion.”

Same AI.
Very different results.

[SGANT @sgantavande]

The Rule:
After a stop-out, I don't touch that name for 90 days.
Not because the stock is bad. Because I am compromised on it. The urge to get it back is the single most expensive emotion in the market, and it doesn't announce itself — it arrives dressed as "but the thesis is still intact."
The blackout isn't about the chart. It's about the fact that I can't be objective about a name that just took money from me.
90 days. No exceptions. It's a calendar entry, not a judgement call.

[SGANT @sgantavande]

I repriced my home loan this year.
Outstanding roughly [S$203k].
New rate: 1.700% fixed, three years. Admin fee waived. Out of pocket: zero.

The broker who helped most recommended against her own commission — she pointed me at repricing with my existing bank rather than a refinance she'd have earned on.

I mention it because people chase 2% edges in the market while leaving a much larger, much more certain one sitting in a loan document they haven't opened since signing.

Check yours before you check your portfolio.

[SGANT @sgantavande]

In July I made RM2,469 on one trade. I still log it as a failure.

Not because of the money. Because of why the money was smaller than it should have been.

Here's the whole thing.

The setup. Sunway Construction (5263.KL) had been building a base for weeks. Pivot at RM7.52.

On 10 July it broke out on 2.05× average volume — the volume confirmation is the part most people skip.

I bought 7,000 shares at RM7.73, then added 5,700 at RM7.80 as it held.

The plan.
Before I entered, I wrote down where I'd sell the add tranche. A ladder: partial out at 8.37, the rest at 8.11 if it faded. Written down. Before the emotion existed.

The mistake. Price ran and I moved my own goalpost to RM8.40. It never printed. I sat there watching a rung that only existed in my head while the stock came back to me.

I eventually filled the ladder as originally written — but late, and 26 sen worse on the second tranche than the first.

What that cost. Roughly RM780 on 3,000 shares. Not ruinous.
But it was a free RM780 that I handed back to the market for the privilege of being greedy for ninety minutes.

The rule I now run. A pre-set floor is not a suggestion. If I want to change an exit, I change it before the setup is live, not while I'm staring at the tape.
The plan written by the calm version of me outranks the plan invented by the excited version of me.

Why this matters more than the RM2,469. I work full-time.— I can't watch a screen at 3pm.
Every rule I write exists to let me trade without being present. The moment I improvise, the whole system stops working for someone with my schedule.

The money was never the lesson. The 26 sen was.

[SGANT @sgantavande]

🚨 Good morning, crypto!
Markets reward patience more often than prediction.
Today I'm watching: • 🟠 Bitcoin trend • 🔵 Ethereum ecosystem • 🟢 Institutional adoption
What's the one coin on your watchlist today?
#Bitcoin #Crypto #Ethereum

[SGANT @sgantavande]

🧵 5 Crypto Mistakes That Keep Beginners Poor (And How to Avoid Them)

Save this before your next trade. 👇

#Bitcoin #Crypto

1/7

🚫 Mistake #1: Buying the hype instead of doing your own research (DYOR).

If everyone on social media is talking about a coin, the biggest gains may already be gone.

✅ Learn what the project does, who built it, how tokens are distributed, and whether there is real adoption before investing.

---

2/7

💰 Mistake #2: Investing money you can't afford to lose.

Crypto is one of the most volatile asset classes. Even strong projects can experience large price swings.

✅ Only invest money that fits your financial plan and won't affect your essential expenses if prices fall.

---

3/7

🎯 Mistake #3: Trying to time the market perfectly.

No one consistently buys the exact bottom or sells the exact top.

✅ Consider Dollar-Cost Averaging (DCA)—investing a fixed amount at regular intervals—to reduce the impact of short-term price volatility.

---

4/7

📈 Mistake #4: Chasing every new "100×" coin.

Most new tokens never become long-term winners.

Meanwhile, projects with strong fundamentals like Bitcoin and Ethereum have shown greater resilience over multiple market cycles.

✅ Build a solid foundation before taking higher-risk positions.

---

5/7

😨 Mistake #5: Letting emotions control your decisions.

Fear can make you sell during downturns.

Greed can make you buy after big price rallies.

✅ Create an investment plan before entering a trade—and stick to it instead of reacting to every headline.

---

6/7

🛡️ Successful crypto investing isn't about making one lucky trade.

It's about:
• Managing risk
• Staying patient
• Continuing to learn
• Thinking long term

The goal is to protect your capital so you can stay in the market for future opportunities.

---

7/7

The investors who build wealth over time usually aren't the ones making the most trades.

They're the ones making better decisions consistently.

Which of these mistakes have you made before—and what did you learn from it?

👇 Let's help new investors avoid expensive lessons.

#Bitcoin #Ethereum #Crypto #Investing

[SGANT @sgantavande]

📊 If you could only hold ONE for the next 5 years...
🔘 Bitcoin (BTC)
🔘 Ethereum (ETH)
🔘 Solana (SOL)
🔘 Other (Comment 👇)
#Bitcoin #Crypto #Investing

[SGANT @sgantavande]

🚨 #Bitcoin is holding above key support while macro uncertainty keeps traders on edge.
The biggest winners in crypto are often those who stay patient when everyone else is emotional.
👀 Watching BTC, ETH, and institutional flows closely. The next major move could surprise both bulls and bears.
What's your outlook for the next 30 days? 📈📉
#BTC #Ethereum #Crypto #Altcoins #Blockchain #Investing

[Brad Garlinghouse @bgarlinghouse]

Over a decade in the making, under a minute to explain. $XRP

[Brad Garlinghouse @bgarlinghouse]

Clear validation of @Ripple Prime’s strength, reliability and tech with today’s investment grade issuer rating from Kroll. Momentum builds when markets recognize these things.

[Martin A. Armstrong @ArmstrongEcon]

https://t.co/2ofvjbBma2

[SGANT @sgantavande]

@ArmstrongEcon What should we buy before and after major low at Aug 2028?

[Brad Garlinghouse @bgarlinghouse]

3 continents, 4 global office visits, 5 days. Crossed too many time zones to count...
Recently, @MonicaLongSF and I (along with others on the Ripple leadership team) traveled to Dublin, London, Singapore and Sydney to meet with the Ripple Team (many of whom joined from our acquisitions of GTreasury, Hidden Road, Rail, Palisade and Solvexia).

A few notes from the road -
1/ centers of gravity (business and/or employee) are never stagnant, and getting out of the US coastal mindset is imperative. It was incredibly energizing to hear from new and longtime Ripplers on what moves the needle where they are.

2/ Culture cannot be taken for granted. More than ever, we’re championing a maniacal focus and eliminating bureaucracy for employees to be owners. Don’t confuse activity with progress.

3/ Adoption doesn’t happen overnight. Platforms > point solutions. Meet customers where they are, not where they might be in a couple years.

4/ AI is becoming a fundamental part of our products – especially in cash forecasting and liquidity management in real-time for the office of the CFO. Employee productivity may be where AI starts, but the end goal is much bigger.

5/ 2026 is shaping up to be another defining year. We’re in the right markets with the right capabilities across payments, custody, liquidity and treasury management. There's a huge opportunity ahead, and we are making sure XRP is at the center of it.