//
sign in
Profile
by @danabra.mov
Profile
by @dansshadow.bsky.social
Profile
by @jimpick.com
AviHandle
by @danabra.mov
AviHandle
by @dansshadow.bsky.social
AviHandle
by @katherine.computer
EventsList
by @katherine.computer
ProfileHeader
by @dansshadow.bsky.social
ProfileHeader
by @danabra.mov
ProfileMedia
by @danabra.mov
ProfilePlays
by @danabra.mov
ProfilePosts
by @danabra.mov
ProfilePosts
by @dansshadow.bsky.social
ProfileReplies
by @danabra.mov
Record
by @atsui.org
Skircle
by @danabra.mov
StreamPlacePlaylist
by @katherine.computer
+ new component
Profile









Loading...
This week on my substack: what is an agent open.substack.com/pub/tidydesi....
This week's productive procrastination involved getting nerd-sniped by seeing if I could get real-time speech-to-text working in Shiny with the Speechmatics realtime API (my speechmatics #rstats 📦 published today on CRAN currently supports their batch API). This was fun though needs lots of work!
Please don't reply to this if you have some product you're selling that allegedly solves this problem - I want to discuss process and methodology
Love this blog post! I've got something drafted to publish next week about how I try to use AI responsibly in my dev work, but this post sums up the main thread perfectly! #opensource #ai tombedor.dev/human-attent...
While many of the big LLM providers can do speech-to-text, I've enjoyed using this dedicated provider for transforming my voice notes into transcripts, and so wrapped a useful subset of their API in an #rstats package, which is now available on CRAN 🎉 cran.r-project.org/web/packages...
2) The triage process is weird as the simplest way is to skip "does this replicate" and assume it does and look for explanations for it which ends up figuring out a fix as well so it's a lot of condensed info, making item 1) worse, and
Has anyone had much success in using AI for auto-triage on #opensource projects? I have plenty of success with using it interactively for this, but automatic never works for me because... 🧵
1) I need to check the output and reading/understanding the explanation is cognitive load so I'm the bottleneck anyway, and so doing it one thing at a time instead of batching is simpler, and
3) I've started small experiments to automatically set up infrastructure to test if things replicate etc but it always feels like more effort than just triaging things myself so these efforts always get abandoned in favour of getting more productive things done (i.e. manually doing it)
Extremely proud and excited that @ropensci.org's community manager @yabellini.bsky.social has been selected for a @sovereign.tech fellowship! It's a much-deserved recognition of her work making #rstats open to all, and a key endorsement of community-building as a fundamental pillar of open-source.
7d
7h
16h