ai, software dev, cyber security and whatever else
Joined March 2016
- Tweets14.3K
- Following3.3K
- Followers5.6K
- Likes23.5K
Pinned Tweet
ever since the twitter algo dropped I keep getting comments on my pinned post.
so if you are an LLM / AI / BOT or otherwise, please say so I can block you.
this suddenly appeared in my gmail
no buttons or instructions or cta no nothing
what is it that you want from me google... just make it go away
testing various theme colors, claude was smart enough to see that modifying the html is wonky and instead rewrote the html in a modular way so iterating color themes is much easier!
it actually saved me tokens and time
𝗔𝗦𝗜 𝗶𝘀 𝗵𝗲𝗿𝗲 - 𝗶𝘁'𝘀 𝗷𝘂𝘀𝘁 𝗹𝗼𝘄 𝗼𝗻 𝗯𝗮𝘁𝘁𝗲𝗿𝗶𝗲𝘀.
---
the thing with AGI is the bar keeps getting raised
yesterday's AGI is today's mediocre LLM
most models today are capable answering virtually any question at PhD level
that within itself I would call AGI
when you factor in the fact it can do that for multidisciplinary fields simultaneously - that should be called ASI
let it ponder the questions of the universe and it will find answers (eg. multiple solutions to Erdos problems)
as long as you let the models run (ie. pay for tokens) they will continue to solve problems
ASI is here. the only reason we don't call it that yet is it's (thus far) low on batteries.
"roast my design then apply changes accordingly"
did not work the first time
but holy shit did it do wonders the second time!
try it!
one of the highest value prompts I've done
they always talk about finding alien signals
but what signals are we transmitting for aliens to find us?
if finding aliens is so difficult, how are we helping others find us?
𝗜 𝗱𝗼𝗻'𝘁 𝗿𝗲𝘃𝗶𝗲𝘄 𝗺𝘆 𝗰𝗼𝗱𝗲 𝗮𝗻𝘆𝗺𝗼𝗿𝗲.
---
code is like assembly now.
in the olden days (pre-LLMs) we would write high-level code.
the entire layer was high-level code and rarely would we ever need to peak into the assembly:
writing, debugging, architecting, reviewing, testing - all were done in the high-level language layer.
---
welcome to present day:
since we don't write code - we write intents, we also shouldn't review code either - we should review intents.
I don't review my code anymore. I ask the agent to generate markdown docs, graphviz diagrams, changelogs, audit reports, etc. I only review that.
I also ask it to write test and evaluate by whether the tests passed or not. I don't need to peak into the tests code - I can also ask plain english, pseudocode, control flow graph, whatever it is I want.
I can ask it to find errors or missing tests and improve that too!
code is like assembly now.
rare are the cases you would need to peak into that level.
𝙒𝙀𝙇𝘾𝙊𝙈𝙀 𝙏𝙊 𝙏𝙃𝙀 𝙁𝙐𝙏𝙐𝙍𝙀!
can you stop saying "one-shot"
none of what the ai doing nowadays is the output to a prompt
it's all reasoning and self-correcting
one-shot = the first actual output from the model's weights
when everyone digs for gold this guy sells the dirt they dig in 🤠
day 12 slow in terms code
but otherwise quite productive
several meetings, calls, setup website, worked on videos
building a business is way more than code unfortunately 🥺
ai discourse be like
2006: "incoherent nonsense that is completely useless"
2007: "waste hours of development time"
2008: "complete slop that feels hollow"
2009: "just an ugly demo that won't work on *real* life scenarios"
