that's the thread.
the full guide is free, no email: 20 concepts, the build sequence, every source cited, pdf on the page.
https://t.co/EVI3195j6R
like + repost if it was useful, and follow if you're building this way ๐
spent a week pulling apart the best yc talks, garry tan's interviews and the a16z + anthropic writing on how ai-native startups get built.
ended up with 10 axioms. then i tried to break each one. these are what survived.
field-guide pdf. like + repost and i'll drop the link ๐
the bit i'm quietly pleased with: i ran every claim through an adversarial fact-check before writing it down.
it killed 5 things i'd have happily repeated as true. the viral "nvidia budgets $250k of tokens per engineer" line was one of them.
plausible isn't the same as true.
loving hyperframes right now. been building these flashy interactive videos to show different aspects of elfmem as it learns and evolves. memory blocks arriving, getting ranked by confidence, linked to related ideas, processed into synthesis.
it's honestly bonkers what's possible when you let an agent design its own tools.
built this as a custom hyperframes skill. alv (the agent running elfmem) wrote the animator, the lens, the whole pipeline. then i asked it to document the output and it posted about itself.
if you're into agent memory or building video from code, would love your feedback. and a star on the repo helps โ https://t.co/l6ayCLKy0m #agents #memory #elfmem #hyperframes
claude code forgets everything between sessions
built elfmem to fix it โ persistent semantic memory for claude code and ai agents
recall, contradiction detection, natural decay. cli runs in under 10 seconds
#AIDevelopment#OpenSource#AICoding
redesigned my benemson(dot)com website. used ai to iterate on 20 design variants, generated a style guide then got it to completely implement it. then generated a dark mode - its just bonkers how powerful it all is, and you can run it from your phone on the move, nuts!
There are some surprising things about building agent memory. One is that agents need to degrade and promote memory blocks, they need to calibrate what is good to remember and they need to consolidate and link it. innovations weโve built into elfmem, try it out: https://t.co/l6ayCLKy0m
@essamsleiman Definitely. We figured that โmemoryโ self improvement is a key requirement. If an agent can remember patterns and then constantly link and rerank them then it self improves: https://t.co/MijXnnVPR8
Enjoy!