Pinned
Dylan Lamb
4,002 posts
Dylan Lamb
@bydylanlamb
GTM. Cold Outbound. Marketing.
London, UK
Joined April 2010
once you understand how gpt works with context windows, token probabilities, and hardcoded system prompts, you stop mistaking prompt theatre for engineering prompt theatre (overcomplicated scaffolding) often looks impressive but doesn’t improve model performance in meaningfulthis ChatGPT prompt went so wildly viral on Reddit. The creator claims to have created this after struggling through 147 failed attempts. basically the prompt flips the usual flow by making the model interview the user first, asking a few targeted questions about purpose,
- been deep diving into likely trends that will come up in 2026. here’s what’s coming fast and underpriced. lock in: 1. mcp-native apps: apps that speak model context protocol out of the box. zero glue code. zero wrappers. agents just work. see langchain, and also what the team
- Replying to @TIMEWhere the F is Demis? Google DeepMind are at the forefront of AI innovation… Kaparthy missing also…? Who made this list? How do you have a job at Time?
- Replying to @the_yanco and @ai_for_successWhoever made this video completely missed the point he was trying to make.
- Replying to @theoThey hate him because he represents a side of them which they are afraid of. Prime’s story is pretty inspiring. He grabbed life by the balls and demanded better for himself then crawled out of a dark place. Most of his haters are too shook to face their demons and lack the
- Replying to @nise_yoshimi and @not_airspeedleave you basement and go explore you dafty
- This wasn’t AI generated. No prompt engineering. No diffusion models. Guerrilla Tea by Pierre Sernet
- Replying to @zoinkHe was also suffered from neuroticism, paranoia and depression. The reason why he exhibited these traits are for these reasons, not because he was “locked in”, though he was obsessed with his work and life purpose. If you purposely behave this way to appear like you’re locked in
- What he really means: “The ai bubble burst is on the horizon. Our app is churning users because they’ve realised it’s vapourware with inaccurate readings. So we will milk the remaining of this cycle with a $1000 course teaching you how to build apps using ai and sell them using"Why sell a course if you’re already making $3.6M/mo?" For a while we have wanted to pursue building our personal brands, however, we never had the economic incentive to do so. Spending an hour on content and education would always make less sense than to spend that same hour on
