streaming the first rebase game https://x.com/i/broadcasts/1pKdRRgamMLJW

so obviously n=1 but anecdotes matter a lot here i gave kimi k3 and sol the same task - simple issue with hovers in the tui being the wrong color sol found and fixed the issue with $0.30 of spend kimi got up to $1.00 and started reading my database before i interrupted it

i did a large change to our keymap system i then asked opencode to try every single keymap to verify it still works and it found all the issues

should we give our slackbot vacation time?

i am evaluating all sandbox providers, our requirements are high - fast disk, not networked - fast CPU, should be virtualizing on bare metal - can pause/resume

i'll be streaming the first game of rebase tomorrow at 3PM EST the lawyers said we had to disqualify our own employees from playing so i'll be spending most of the time roasting ryan

OpenTUI is designed to attract angry idiots - TUIs make people mad - javascript makes people mad - it's also zig which makes people mad

i regret to inform all the people who get mad for some reason whenever people build a TUI, OpenTUI is at 300,000+ weekly downloads

the next generation of OSS models are quite large it means they're not going to be as cheap to run (will still be cheaper than frontier) will be interesting to see what the gap looks like then

the only valid answer i got to this was wispr flow a lot of people thought granola and sumo were founded in sf but they are not

we're in a time where it feels like every company is being reduced to an api switching products is easier than ever with the help of AI and yet the majority of companies being founded are B2B

my timeline is all mid level employees trying to dunk on each other being like "stop bragging about how hard you work, i work harder but i'd never say it"

prompt injection risk is so annoying man we could do so much cool automation if this wasn't a problem

my team builds this app and they don't even include a head type that represents me

one thing about building products is of course you get people who disagree with your decisions but they frame it larger than a personal preference "change this or you will fail" that's when you know they're definitely wrong

app store review process sucks

please i'm begging you show me something you built not another "this is my custom agent setup" post where you pretend you're doing something smarter than vanilla claude code please

all these static PR review tools are pointless no matter how fancy they are you can tell your robot about a PR, ask it what to look at first and go back and forth with your thoughts until you've processed it all interactivity is always better

slack connect allows us to efficiently collaborate cross functionally with the best teams across the industry

we have an issues channel where every good github issue goes and we can ask opencode to reproduce it and fix even started throwing posts from here in there

opencode2 has a new api that is very very good we included a cli client so you or your agents can query it easily

every person with a take a company's profitability is a moron

whatever tool you use just make sure it's taking advantage of everything models can do they are intensely trained for certain behaviors and they need the right structure around them if you're not seeing much of a difference in model versions you're using the wrong tool

claude code grew so fast because they had the most usable model say something, get a magical result pretty quickly these ultra smart slow expensive models like fable (and the old codex ones) are the exact opposite

new settings screen i prompted slackbot around 5pm today and asked for some concepts it sent screenshots and videos but i was afk kit jumped in and gave it feedback. then he stepped away then i took over back and forth until it was done - it even made this video

what i learned from opentui is investing in primitives that enable prompting is so worthwhile everyone is having fun prompting their own tuis now there's other missing primitives and i think we need to invest there as well

companies get a bunch of GPUs to do model hosting they realize model hosting is hard and they could just rent the GPUs given how much demand there is right now so now there's like 2 companies who who do model hosting well

forgot to do this for a bit june was 13,200,000 monthly actives

guys we have a pretty substantial opensource zig codebase and i'm terrified he's gonna look at it

being in the loop is important but i keep seeing people misunderstand this - your job isn't to micromanage the agent and get it to execute its steps in a way you think is better than what it'd do on its own it's to come up with cool things for it to work on

issue comes in tag opencode opens a PR with a video verifying fix

if the bot has its own computer it can make its own apps

if you want to help beta test OpenCode 2.0 http://v2.opencode.ai - data in separate db which we might wipe - stuff will be broken - use /report to send us issues - v1 plugins won't work, v2 api not final there is a built in skill that you can ask for basically anything

i wish we could all be less serious we should all just be laughing at that post about bun instead of taking it seriously the reason it got written in the first place is because everyone is way too serious about dumb stuff

at some point you have to be embarrassed about having an opinion about everyone and never being the person people have an opinion on

what are best products from ai startups in sf right now actual products that end users use

one fun thing is you don't actually need seperate slack apps for fun little things we asked opencode to keep us updated on world cup and it set it up we have no idea how this works but it just works

if you are around at 2PM EST tomorrow we're doing a test run of something fun we've been working on would need an iPhone, be in the US and 20min if you're in can you drop a reply in this thread

this will be useful

multimodal sensory prompt fusion

one thing i am worried about is that it's possible that 50% of our enjoyment came from its speed and maybe it doesn't continue to be as fast once it's publicly available

i've never hyped a model release, we're generally conservative with how we use these things but gpt-5.6 has had a massive impact on our team, we're using 5x the tokens as we used to it's not even smarter than fable or anything, but it's just so reliable and fun to use

the way the anthropic team talked the impact of their slack bot was very similar to our own experiences so i assumed the products would be very similar 1/3

there's infinite talk about ai taking jobs but i hardly see people talking about it disrupting the company that employs them we're tossing out products we've used forever it's not a price thing, we just get a better experience by hooking ai up to raw data

what docs service actually has good design? the ones i've looked at have LLM quality design if a company is only going to focus on docs i imagine one of them must have killed it on the design front?

imagine if LLMs were invented any any previous point - say 20 years ago when a bunch of tools it loves to use didn't even exist do you think it would have accelerated them existing? or would it have calcified things

we have customers who we are customers of i see money coming in that we then just take and pay back is the economy even real

erlang haaland

guys stop trying to argue with the "code doesn't matter" posts and just ask them if they're so ahead of the game why aren't they richer

i feel like every debate on here breaks down into employed vs unemployed