AI Educator. Helping you to your AI, Tech, Tools & Digital Skills With Multi.💬 DM for collaboration✉️codernihan@gmail.com

Earth
Filter
Exclude
Time range
-
Minimum likes
Nihan retweeted
Most people open ChatGPT, Claude, Gemini or Others and type 3 words. "Tell me." "Do this." "Write this." Then they get a safe, average answer and blame the model. The model was never the weak part. The first line was. That first line decides who answers, what "done" looks like, and what it is not allowed to say. Miss it, and every reply after that stays flat. No matter which model you paid for. 5 rules below. The 2nd one shows why your last 10 prompts felt empty. The 4th one is the line that stops the confident bluff. Read rule 1 before you send another prompt.
8
12
19
662
Save this. 1. Role first, then limits 2. Goal = the reader's state, plus what not to say 3. One step at a time 4. Force unsure facts to be labeled 5. Make it cut its own draft, then rewrite Next time, match the ask to these 5 before you hit send. Don't send "tell me" or "do this" alone.
42
Rule 5: Do not accept a 60-point answer. Send this: "Score this answer out of 10. Name 3 weak spots. Rewrite only those. No praise." People who stop after one round get an average draft. People who go three rounds get a different level of draft.
1
42
Rule 4: Add one line that blocks the bluff. "If you are not sure, label it unconfirmed. Do not mix guesses with facts." That line alone stops most confident-sounding filler. Be extra strict on numbers, dates, and names. The checking you do later gets much shorter.
1
23
Rule 3: Never ask for the whole job in one prompt. Plan. Then structure. Then draft. Then cut. Ask for all of that at once, and the middle collapses. One step. Read the output. Then the next step. Rushing is the slow way. This is how people who use these tools well actually work.
1
32
Rule 2: Don't describe the object. Describe the state. Bad: "Write a post." Good: "Write a post that makes the reader feel this is their problem, and want to do one thing right now." Then show it what to avoid. "Never use lines like: 'In today's digital world.' 'Let's find out.'" "No hype tone." A bad example cuts off-target answers fast.
1
40
Rule 1: Decide who answers, in the first line. "Answer as a marketing consultant who works on content for small businesses." That one line changes the path it thinks on. Expert mode, switched on. Then lock the frame. Freedom makes the answer weak. • Under 120 words • No jargon • Order: decision, then reason, then one example "Write it however you want" is the worst instruction you can send. Clear frame. Sharper answer.
1
58
Replying to @ArafatMd93059
Thanks for Sharing
21
Nihan retweeted
My neighbor spent five years working as a night manager for a major global hotel chain. We were talking yesterday and I asked him what the front desk actually sees on their screen when you walk in. He gave me a list of things that made my stomach drop. I am never booking a hotel room the same way again.
7
18
22
681
Nihan retweeted
🚨This is EXTREMELY dangerous. I just watched an 18-year-old guy calmly explain how he makes $10,000 a month using AI while he sleeps, and I'm convinced that 99% of people still aren't seeing this. The whole system: Pick a YouTube channel that posts content frequently. Paste the link into a tool. Connect your TikTok, Instagram, and YouTube Shorts accounts. Close the laptop. From there, the AI does everything. It detects every new video that's posted, finds and clips the moments with the highest viral potential, adds subtitles, and distributes them automatically across all platforms. 1 million views ≈ $2,000 Zero editing. Zero manual posting. Zero stress. 10 minutes to set it up and then it runs on autopilot. You're no longer creating content. You're building a content factory. Save it.
7
11
13
532
I AM QUITTING MY JOB TO GO FULL IN CLAUDE Just asked him to: "Analyze misspriced Polymarket markets opportunities for arbitrage and find wallets that are using it to copy" Turned $2K into $12K in one night Monitored ~1k+ wallets I just realized that there are many arbitrage bots that I can't beat without code knowledge But I can find them and copy So Claude created a monitoring terminal and copytraded found wallets using TG copytrading bot It's not a script and not even the bot, it's an AI agent that is improving with each found wallet Fetching wallet behaviour, how it's trading, arbitrage, what's sized and timings 70% win rate, 7 wallets copytrading rn from ~500 monitored, bot never paused, never gambling, just math and profit Giving This Free for 24 hours. To get it: 1. Comment the word 'Claude' 2. Like and Retweet this post 3. Follow me @tec_marco10 (so i can DM you)
108
84
212
14,278
Replying to @Ecarter_ai
Send
2
Replying to @ThatsEFM
Well Done
26
Replying to @Acarter_coder
Excellent Share
4
Replying to @sufyanmaan
Excellent
64
Replying to @Aura_lixx
Awesome
6
Replying to @Arina_sharmi
Awesome
3
Replying to @Seok_Ai0
Awesome
5
Replying to @TCibelle
Love this
5