Andrew Lee
speaker
2,833 appearances
3 recordings
1 series
first heard Mar 2025
last heard 15 May
Andrew Lee’s voice in public audio — every appearance, attributed to the second.
Trend
recordings per month · last 12 monthsRecordings per month over the last 12 months — 2 in all, peaking in May 2026 with 1.
Appearances
Um so I we're not we're not using any any other providers because we think this is something that we have to solve ourselves.
Um and if our whole premise is we're like ChatGBT or Claude, but like much more heavyweight and designed for like long running recurring work, like this is this is a very core uh fundamental thing we need to do.
Um the North Star
for us is the illusion that we're just one big long chat and everything has been put into that chat and that the model is perfectly smart.
And um the the system breaks under sort of all of those over all of those reasons, right?
The you know we can't actually fit all of it in context.
We can't afford to do what we could and the model is not infinitely smart.
So if you you know for example you stored every email that you'd ever process just in the history um the
L O M isn't gonna do a good job of reasoning about which ones you processed or which ones you haven't processed.
It's just
It's just not gonna be that smart.
Um and so one of the techniques that we've been using is having a data structure that is managed explicitly by the LLM across runs.
Um the first iteration of this was a very simple version, which was in the system prompt, we included a JSON blob.
And that JSON blob could be directly edited by a tool called in the LLM.
So in the subagent runs, you could just tell it, like, hey, I want to add this data, remove this data from
Um this dataset object.
And that actually worked surprisingly well.
This was like the dumbest thing we could think of.
And we built it and it just worked and it was really cool.
Um, but it came with a couple of downsides.
Showing 1141–1160 of 2,833 · page 58 of 142
← Previous
Next →