Post

im not kidding and im sure @cameron.stream can back me up from another direction: if you arent capturing/mining your team's agent session logs, you are leaving it all on the table. we have a service that lets you run RAG tool calls against vectorized agent session summaries and it is 🔥🔥🔥

austin (eeek!/ackkk! 👻)@thebadcode.com

the fact that ant/oai dont trivially expose a "download all my org's inference logs" button makes me think that they realize those logs are worth almost as much as the output of the agent sessions themselves

26/7/26, 9:27 pm5 reposts79 likes

brennan

@brennan.computer

27 July

yeah, moving to having everything in the open and then iterating and looping that into itself is a big bet here: shopify.engineering/under-the-ri...

shopify.engineeringUnder the River (2026) - ShopifyWhat it took to ship our Slack-native agent River, lessons learned, and the substrate that runs beneath it. Co-authored by River.

Cameron

@cameron.stream

26 July

One of our users has the line "if it didn't happen in Letta, it didn't happen"

atticus goldfinch

@atticusgf.bsky.social

26 July

Can you say more about your tool? Use cases, etc?

austin (eeek!/ackkk! 👻)

@thebadcode.com

26 July

RAG on claude sessions:

skelly tayne

@ohnoether.bsky.social

26 July

you find RAG to be worth it for this sort of things over just agent grep? i still default to the lazy thing here, but mostly out of ignorance

austin (eeek!/ackkk! 👻)

@thebadcode.com

26 July

one nice thing about RAG is that it forces you away from thinking in terms of "i will grep my local logs" and into "im building an online system that many useful things will dump events to" thinking