“i was never going to be limited by hoodies and t-shirts… no matter how much it made sense” - virgil abloh

great question. there were a few ways to manage this but we settled on managing drift/version control semantically and backing it by a git repo. - whenever a change is made to any specific change is made to a dimension (product, company, etc) a log file is created that has more details on why, source/references, etc. - there's a changes.jsonl file that tracks everything - its backed by git so you can see when teammates make changes

just launched a free/open source app that makes your entire team's agents better (link below) it is a desktop/cli app that: - grounds each session with context so you don’t have to focus on steering your agent with nuances of your product, team, etc - can connect to slack, github, and granola so your context stays up to date with out any extra work on your part - you can push/load context to your team to collaborate on agent context - works across different agents (claude code/cowork, codex, openclaw, etc). - works with existing usage so there’s no extra fees to be paid - backed my local files so you can edit/take files with you and more!

just launched a free/open source app that makes your entire team's agents better (link below) it is a desktop app/cli that: - grounds each session with context so you don’t have to focus on steering your agent with nuances of your product, team, etc - can connect to slack, github, and granola so your context stays up to date with out any extra work on your part - you can push/load context to your team to collaborate on agent context - works across different agents (claude code/cowork, codex, openclaw, etc). - works with existing usage so there’s no extra fees to be paid - backed my local files so you can edit/take files with you and more...

"you don't have to predict the future if you can see the present moment clearly."

you can't run from reality.

trying to write for 30 uninterrupted minutes really exposed how cooked my attention span really is

launch in 9 days. time to put belt to ass.

reading sam presti press conference transcripts like they are warren buffet annual letters

it gets real interesting. stale context gets updated by a subagent that the main agent calls autonomously or that you could explicitly call to update something. agents stepping on each other is still tricky*. right now, it still takes a bit of manual curation and is best either having subfolders per person (tech leads own `./engineering/`, pm owns `./product/` etc.) or having someone take point on keeping things clean. all changes are logged in a changes.jsonl file so there's explicit paper trail and you can have an agent manage overwrites/bad merge conflicts much the same way you could for code. *but i'm also working on something new (and of course open source) that will make this much more seamless.

open source plugin for shared context across your entire team and all of your coding agents

released/open sourced a plugin that has been helping me build and work with teammates for a while

built a claude code/codex plugin that keeps product context consistent across sessions and llm cli agents. github link below

i built a claude code/codex plugin that keeps product context consistent across sessions and llm cli agents. github link below

people seem to think as coding tools get better that they have to "know" less as an engineer. and i don't think that's true. or at least not the right framing for what you do and don't need to know.

its all about execution.

leon thomas was right the vibes really don’t lie