Bren

a trend im seeing amongst teams. if your AI coding assistant keeps "forgetting" your codebase, the fix usually isnt another tool. its telling it where things are. before you install a "give your agent a memory of your whole codebase" plugins, know this: the people who build AI coding tools already tried that and backed away from it. early versions of claude code indexed your code into a search database. anthropic ripped that out and did the simple thing instead: let the AI search and read your files the way you would. sourcegraph builds code search for a living. they tried the index too, then dropped it and went back to plain keyword search. so treat the index as guilty until proven necessary. you really only need it when BOTH are true: 1) your codebase is huge 2) you keep asking things plain search can't answer, like "what calls this function" before that, KISS and use what already works. keep a short notes file the AI reads so it knows how your project is laid out. use plain find files for "where does this live." karpathy does similar for his own knowledge: he keeps it in plain markdown files the AI reads, because the upload all and search setup makes the model relearn everything from scratch every time. he's talking about personal notes, but it's the same move for code. one real exception: when you don't even know what something is called, a meaning based search helps you find it. and the promblem with an index, its a point in time snapshot. right after you change code it quietly points the AI at the old version. so before you add the tool, just ask the AI how your project works.

yea, one way or another

@PolymarketMoney **study funded by data center company

JUST IN: New study shows data centers has reduced the cost of electricity in America.

@AgentsDispatch planning one for coding another for review another for QA/TESTING

@BrenBuilds You should never use just one model (even Fable). Using multiple, simultaneously on the same problem in a cooperative or adversarial way yields results an order of magnitude better.

"im a claude code user should I switch to codex or cursor?" "I'm a codex user should I switch to claude code?" Don't be so tribal and black or white. Use everything at the top of product lines.

We're going to get GPT 5.6 before neutered Fable 5 rerelease aren't we

@danielendara Copilot

@BrenBuilds just curious if you can share... what enterprise solution did they go for.

talked with a company recently, $25 a month is what their 500 developers get in token usage a month. they had the audacity to say "we aren't seeing the impact of AI." brother, at that spend, you WONT.

@johnhelmuth_ True. It's becoming more expensive weekly

@BrenBuilds Hard to do that everything costs hundreds of dollars monthly. Most people can only pick 1.

@clairevo I'm not sure what the output even is at that price point

@BrenBuilds Nooooo

@pouriaa the pace of release is so fast right now

@BrenBuilds Keep switching between them every few months since the product velocities from these companies is so high. Even if you hate <TOOL> today, they might ship a feature that makes you love it tomorrow

Mass exhodus from Google DeepMind, so it's over for them. It's Anthropic vs OpenAI and the latter has way more market share

@theo i left it with a $10k MRR goal last week, must be close to finishing. Will check tomorrow.

3 days left of using Fable in your Claude Code sub! Better maximize that token usage

@theo I had plans to take over the world with it this weekend.

I won't lie, really thought we'd have Fable back by now. Didn't think we'd go over a week.

Find yourself someone who loves you as much as gpt 5.5 xhigh loves making tests

@alfondotnet @benln @RevenueCat 😍 agreed

@BrenBuilds @benln @RevenueCat unlimited, obviously we don’t like wasting just because

Which startups are actively hiring? Reply with company name, career page, location.

a trend im seeing amongst teams. if your AI coding assistant keeps "forgetting" your codebase, the fix usually isnt another tool. its telling it where things are. before you install a "give your agent a memory of your whole codebase" plugins, know this: the people who build AI coding tools already tried that and backed away from it. early versions of claude code indexed your code into a search database. anthropic ripped that out and did the simple thing instead: let the AI search and read your files the way you would. sourcegraph builds code search for a living. they tried the index too, then dropped it and went back to plain keyword search. so treat the index as guilty until proven necessary. you really only need it when BOTH are true: 1) your codebase is huge 2) you keep asking things plain search can't answer, like "what calls this function" before that, KISS and use what already works. keep a short notes file the AI reads so it knows how your project is laid out. use plain find files for "where does this live." karpathy does similar for his own knowledge: he keeps it in plain markdown files the AI reads, because the upload all and search setup makes the model relearn everything from scratch every time. he's talking about personal notes, but it's the same move for code. one real exception: when you don't even know what something is called, a meaning based search helps you find it. and the promblem with an index, its a point in time snapshot. right after you change code it quietly points the AI at the old version. so before you add the tool, just ask the AI how your project works.

@BenjDicken Love this. 2 hours a day wasted on no outcome status meetings. Working on shedding.

TLDR

@AGI_Pulse @jietang @elonmusk @teortaxesTex It's a trap 🪤

@BrenBuilds @jietang @elonmusk @teortaxesTex Don't even ask my bro

I think GLM 5.2 points to a 7 months gap currently It's around Opus 4.7-4.8 level, all told (modulo vision which in Opus's case is garbage anyway). Mythos reached Preview status (≥ Opus 4.8, functionally) by early Feb 2026. This means full PRC Mythos ("Fable") by Nov-Dec'26.