this is f*cking gold
the Claude setup most people will never find on their own
if I had this a year ago, I would've worked 5x faster
in the right hands, this changes everything:
Uber partnering with Expedia on hotels gets to validate hotel booking intent inside its ecosystem without owning supply. But if it works, why keep Expedia in the middle
https://t.co/fILByIinup
@andruyeung@jappleby I’ve seen initiation fee up to $5k and annual fees $10k. This seems more mid market than other clubs in NY. Interesting to see growing niche private clubs because there are so many now that they need to differentiate for members
Favorite tool to store and manage prompt library? Notion/Excel/Doc all seem outdated and hard to search solutions.
Is an app the solution? Typingmind or Droppromot?
Big news for NY startups! QSBS is here to stay.
From @politico: the idea now “seems to be moot” following strong, coordinated engagement from across the tech community—including a @TechNYC letter with 1,600+ founders, early employees, and investors.
This is a clear example of what’s possible when the ecosystem shows up together. We’re grateful to everyone who spoke out and helped ensure policymakers understood what was at stake.
New York remains the best place to build. 🗽
Prompt engineering isn’t dead…What survived is the discipline of turning professional judgment into precise instructions, then transmuting those instructions over time into reusable intelligence, compounding in the reinforcement learning environment of the white collar workplace
@TukiFromKL I can’t decide if bankers agreeing to only $100/hr to train Grok to replace themselves is more jaw dropping or Open AI cloud selling to Amazon
How is it only Wednesday?
@Onil_coder What advantages do prompt generators have over asking LLMs directly to help refine or produce a new prompt? Especially when prompt structures seem to be so specific to the model?
Love this breakdown of how to uniquely structure prompts based on AI tool because AI models are not one size fits all. I’ve found that I can get very different responses with even the same prompt across tools!
Stop using the same prompts across different AI tools.
Smart users customize their approach for each tool.
Here’s why learning each model’s style matters:
[ 🔖 bookmark this post for later ]
1. ChatGPT delivers when you guide it like an instructor - asking for step-by-step breakdowns, practical examples, and setting a clear, concise tone.
2. Perplexity excels when treated like a research analyst - ask for specific facts, request sources and citations, and include relevant keywords with timeframes.
3. Grok responds best when you get candid - think raw feedback, emotional angles, skip the corporate jargon, and ask for quick, tweet-style insights.
4. Gemini works well when given structure like a project planner - ask it to organize content, define clear objectives, and format everything into workspace-compatible outputs.
AI models aren’t one-size-fits-all.
Tailor your prompts to the specific model for better outputs.
Save this guide. Share it. Start with one AI tool.
📌 Get Advanced ChatGPT Guide (free): https://t.co/kOBWfKrBaX
👉 Follow me @AndrewBolis for more and 🔄 Repost this to help others use AI
@heyrimsha are there prompts you use to generate an even better prompt? Lately I’ve been asking ChatGPT/Claude to refine my prompt for increased clarity + deliverables even after inputting a structured prompt: goal > context/ref > deliverable > constraints/risks > alignment+plan
The all-too-familiar marketplace supply growth <> quality control problem. In a race to penetrate new markets and meet demand, marketplace startups in a saturated comp pool too often sacrifice onboarding. Home service startups are prime examples of this. https://t.co/I3u7Hf2HWi