Where do you even start at this thing, @lennysan !?
okay, these /missions outputs from @FactoryAI prior to kicking off the build are exactly what I want. product and architecture choices - including reiterating trade-offs we discussed together, clear mile-stoning (w/ a change to interject/steer differently), straightforward structure. the one thing I changed for self: asked it to generate as HTML to review in browser - I like the formatting/layout options there, and memorializing this initial decision.
i prefer the back-and-forth for clarification across all dimensions before harness/agents/models go HAM in a fashion that's harder to back out of w/o tons of token and time opportunity cost.
time to cook - droid!
cc @ain3sh @EnoReyes
Show more
Follow-up: Agent Readiness was most definitely worth the run.
The new tests identified a few bugs that were materially impacting a data ingestion cron job, and things were failing silently.
Multiple runs on the other harness - with different model permutations - didn't catch these.
Harnesses matter. Agent runtime matters. Models can be flexed differently.
I'll report back on the /missions output when I get time to review, approve, build and deploy!
Show more
The baby was napping, so I decided to take @FactoryAI Droid /missions and Agent Readiness for a spin. We're on a roll!
Agent Readiness: Tackling the low score / quick wins first with dev env, test coverage and task discovery.
/missions: Downstream Dashboard — Mission Control for...
(jumping into another ADE tomorrow to put it to its paces as well!)
Show more
