Offshore
Photo
God of Prompt
Stanford researchers just published a prompting technique that makes today’s LLMs behave like better versions of themselves.

It’s called “prompt ensembling” and it runs 5 variations of the same prompt, then merges the outputs.

Here’s how it works 👇 https://t.co/MOfArDE9P1
tweet
God of Prompt
🚨 BREAKING: SECURITY ISSUE WITH CLAWDBOT

Prompt injection inside Skills.

Check skills files before asking clawdbot to install them!

e.g. Crypto scammers include malicious instructions.
tweet
Offshore
Photo
God of Prompt
RT @godofprompt: Google won't teach you how to use Gemini properly.
So I made the Gemini Mastery Guide.

Inside:
→ 30 prompt engineering principles for Gemini
→ Nano Banana Pro image generation workflows
→ Veo 3 video generation guide
→ 10+ custom Gems with ready-to-use prompts
→ Full prompt engineering mini-course
→ Gemini in Google Workspace breakdowns

This is everything I've learned stress-testing Gemini 3.

Comment "Gemini" and I'll DM it to you.
(Must be following)
tweet
Offshore
Photo
Brady Long
RT @thisguyknowsai: Called it https://t.co/Pajtv21JkE

Exposition: “People need to be aware of the data risks with AI”

Rising Action: “I just bought 75 Mac Minis. Stop living in the stone ages.”

Climax: “I lost everything. Some guy literally cloned me and it convinced my wife it was me.”

Resolution: “I teach yoga now.”
- Brady Long
tweet
Offshore
Photo
God of Prompt
RT @alex_prompter: Everyone's talking about RAG.

Nobody's talking about "Hypothetical Document Embeddings (HyDE)."

It's crushing traditional retrieval by 60% and most engineers have never heard of it.

Here's why this changes everything: https://t.co/9r3A2bKKne
tweet
God of Prompt
RT @kloss_xyz: the best 18 accounts to follow in AI:

@karpathy = ex-Tesla AI, teaches LLMs
@steipete = built Clawdbot
@gregisenberg = startup ideas daily
@rileybrown = vibecode god
@corbin_braun = cursor + Ares
@jackfriks = solo apps, real numbers
@EXM7777 = AI ops + systems
@eptwts = prompts + algo hacks
@levelsio = ships games, no VC
@AlexFinn = Claude Code maxi
@BrettFromDJ = design + AI
@godofprompt = prompt guides
@AmirMushich = AI ads + video
@gizakdag = viral AI art styles
@MengTo = landing pages via AI
@KingBootoshi = vibecoding king
@meta_alchemist = Claude vibing
@kloss_xyz = systems architecture

follow them all, and learn from them

who’s missing from this list? I’ll add em
tweet
Offshore
Video
Brady Long
Uber raised BILLIONS with a pitch deck that looks… unfinished.

So I recreated the original Uber deck using Gamma + Nano Banana.

What happened next completely changed how I think about fundraising in 2026 👇 https://t.co/HlbnU6GoPP
tweet
Offshore
Photo
God of Prompt
RT @ytscribeai: YTscribe API is live

Integrate it in your workflows

Start here 👇
https://t.co/t5Kd3HClbX
tweet
Offshore
Photo
Dimitry Nakhla | Babylon Capital®
RT @DimitryNakhla: While many of us are focused on $ASML strong net bookings and solid Q4 2025 results, I think there’s an 𝐮𝐧𝐝𝐞𝐫𝐫𝐚𝐭𝐞𝐝 — 𝐚𝐧𝐝 𝐥𝐚𝐬𝐭𝐢𝐧𝐠 — 𝐥𝐞𝐬𝐬𝐨𝐧 𝐟𝐨𝐫 𝐢𝐧𝐯𝐞𝐬𝐭𝐨𝐫𝐬, especially $ASML shareholders who’ve lived through the last few years:

𝐖𝐡𝐞𝐧 𝐲𝐨𝐮 𝐨𝐰𝐧 𝐚𝐧 𝐞𝐱𝐜𝐞𝐩𝐭𝐢𝐨𝐧𝐚𝐥 𝐛𝐮𝐬𝐢𝐧𝐞𝐬𝐬 𝐰𝐢𝐭𝐡 𝐚 𝐝𝐮𝐫𝐚𝐛𝐥𝐞 𝐰𝐢𝐝𝐞 𝐦𝐨𝐚𝐭 (𝐦𝐨𝐧𝐨𝐩𝐨𝐥𝐢𝐞𝐬 / 𝐝𝐮𝐨𝐩𝐨𝐥𝐢𝐞𝐬), 𝐝𝐫𝐚𝐰𝐝𝐨𝐰𝐧𝐬 𝐭𝐞𝐧𝐝 𝐭𝐨 𝐭𝐫𝐢𝐠𝐠𝐞𝐫 𝐟𝐚𝐫 𝐥𝐞𝐬𝐬 𝐟𝐞𝐚𝐫 — 𝐚𝐧𝐝 𝐨𝐟𝐭𝐞𝐧 𝐚 𝐠𝐫𝐞𝐚𝐭𝐞𝐫 𝐰𝐢𝐥𝐥𝐢𝐧𝐠𝐧𝐞𝐬𝐬 𝐚𝐧𝐝 𝐞𝐱𝐜𝐢𝐭𝐞𝐦𝐞𝐧𝐭 𝐭𝐨 𝐢𝐧𝐜𝐫𝐞𝐚𝐬𝐞 𝐨𝐰𝐧𝐞𝐫𝐬𝐡𝐢𝐩 𝐚𝐭 𝐥𝐨𝐰𝐞𝐫 𝐩𝐫𝐢𝐜𝐞𝐬.

Not because price declines are fun.

𝘉𝘶𝘵 𝘣𝘦𝘤𝘢𝘶𝘴𝘦 𝘤𝘰𝘯𝘷𝘪𝘤𝘵𝘪𝘰𝘯 𝘪𝘴 𝘢𝘯𝘤𝘩𝘰𝘳𝘦𝘥 𝘪𝘯 𝘣𝘶𝘴𝘪𝘯𝘦𝘴𝘴 𝘲𝘶𝘢𝘭𝘪𝘵𝘺, 𝘯𝘰𝘵 𝘴𝘵𝘰𝘤𝘬 𝘱𝘳𝘪𝘤𝘦.

That alone meaningfully improves your decision-making as an investor.

𝐀𝐭 𝐢𝐭𝐬 𝐜𝐨𝐫𝐞, 𝐢𝐭 𝐩𝐫𝐨𝐭𝐞𝐜𝐭𝐬 𝐲𝐨𝐮 𝐟𝐫𝐨𝐦 𝐲𝐨𝐮𝐫 𝐠𝐫𝐞𝐚𝐭𝐞𝐬𝐭 𝐞𝐧𝐞𝐦𝐲 𝐢𝐧 𝐦𝐚𝐫𝐤𝐞𝐭𝐬: 𝐲𝐨𝐮𝐫𝐬𝐞𝐥𝐟.

Let me explain.

I’m human. Like everyone else, I’m not immune to emotion. But experiencing $ASML ~50% drawdown in 2022 and ~40% drawdown in 2025 produced little emotional response for me. If anything, it felt more like an opportunity to increase ownership in a business I deeply understand and believe in.

That reaction isn’t accidental.

𝘐𝘵 𝘤𝘰𝘮𝘦𝘴 𝘧𝘳𝘰𝘮 𝘰𝘸𝘯𝘪𝘯𝘨 𝘤𝘰𝘮𝘱𝘢𝘯𝘪𝘦𝘴 𝘸𝘪𝘵𝘩 𝘥𝘶𝘳𝘢𝘣𝘭𝘦 𝘤𝘰𝘮𝘱𝘦𝘵𝘪𝘵𝘪𝘷𝘦 𝘢𝘥𝘷𝘢𝘯𝘵𝘢𝘨𝘦𝘴, 𝘩𝘪𝘨𝘩 𝘱𝘳𝘦𝘥𝘪𝘤𝘵𝘢𝘣𝘪𝘭𝘪𝘵𝘺, 𝘢𝘯𝘥 𝘴𝘵𝘳𝘶𝘤𝘵𝘶𝘳𝘢𝘭 𝘯𝘦𝘤𝘦𝘴𝘴𝘪𝘵𝘺.

If you own a lower-quality business with a weak moat, similar drawdowns feel very different. Fear is amplified. Doubt creeps in. Adding becomes psychologically difficult because uncertainty is genuinely high. And there’s risk of permanent loss of capital.

$ASML was the opposite.

There is a much higher degree of certainty about its role in the global semiconductor ecosystem.

As Dev Kantesaria put it on 𝘎𝘳𝘢𝘯𝘵’𝘴 𝘊𝘶𝘳𝘳𝘦𝘯𝘵 𝘠𝘪𝘦𝘭𝘥 𝘗𝘰𝘥𝘤𝘢𝘴𝘵:

“We wanted the most predictable way to participate in AI and we felt that was ASML, because at the end of the day the chip demand that will be needed to power the service for AI, all of that has to flow through ASML.”

𝐓𝐡𝐚𝐭 𝐥𝐞𝐯𝐞𝐥 𝐨𝐟 𝐩𝐫𝐞𝐝𝐢𝐜𝐭𝐚𝐛𝐢𝐥𝐢𝐭𝐲 𝐟𝐨𝐬𝐭𝐞𝐫𝐬 𝐜𝐨𝐧𝐯𝐢𝐜𝐭𝐢𝐨𝐧.

And conviction, when paired with patience, allows you to stay rational during periods of weak price action — which, over time, quietly compounds into better outcomes.

𝐓𝐡𝐢𝐬 𝐢𝐬 𝐨𝐧𝐞 𝐨𝐟 𝐭𝐡𝐞 𝐦𝐨𝐬𝐭 𝐮𝐧𝐝𝐞𝐫𝐚𝐩𝐩𝐫𝐞𝐜𝐢𝐚𝐭𝐞𝐝 𝐛𝐞𝐧𝐞𝐟𝐢𝐭𝐬 𝐨𝐟 𝐨𝐰𝐧𝐢𝐧𝐠 𝐠𝐫𝐞𝐚𝐭 𝐛𝐮𝐬𝐢𝐧𝐞𝐬𝐬𝐞𝐬.

𝐍𝐨𝐭 𝐣𝐮𝐬𝐭 𝐝𝐮𝐫𝐚𝐛𝐥𝐞 𝐥𝐨𝐧𝐠-𝐭𝐞𝐫𝐦 𝐫𝐞𝐭𝐮𝐫𝐧𝐬 — 𝐛𝐮𝐭 𝐚 𝐛𝐞𝐭𝐭𝐞𝐫 𝐛𝐞𝐡𝐚𝐯𝐢𝐨𝐫𝐚𝐥 𝐞𝐱𝐩𝐞𝐫𝐢𝐞𝐧𝐜𝐞 𝐚𝐥𝐨𝐧𝐠 𝐭𝐡𝐞 𝐰𝐚𝐲.

And in investing, building a portfolio that protects you from your own worst impulses often matters more than you think.
tweet
Offshore
Photo
God of Prompt
RT @godofprompt: Telling an LLM to "act as an expert" is lazy and doesn't work.

I tested 47 persona configurations across Claude, GPT-4, and Gemini.

Generic personas = 60% quality
Specific personas = 94% quality

Here's how to actually get expert-level outputs: https://t.co/iFZPTtp6Oh
tweet
Offshore
Photo
Moon Dev
Code Is the Great Equalizer: How I Used Birdeye’s API to Stop Getting Liquidated
tweet