anil.recoil.org
Professor of Planetary Computing at the University of Cambridge @cst.cam.ac.uk, where I co-lead the @eeg.cl.cam.ac.uk, and am also to found at @conservation.cam.ac.uk.
Homepage at https://anil.recoil.org
375 posts
1,260 followers
417 following
Regular Contributor
Active Commenter
comment in response to
post
I’d be absolutely delighted to! (right after the immediate exam week ends). Will drop you an email
comment in response to
post
YAY!!!
comment in response to
post
and hold a conversation with a tree
comment in response to
post
It's hard work because you need to dance like a butterfly
comment in response to
post
en.wikipedia.org/wiki/The_Gon... is excellent!
comment in response to
post
perils of being an early adopter, i guess! :-)
comment in response to
post
wonder who this @avsm.bsky.social is, hello! I didn't realise that ATProto handles could be reused once renamed to another domain like my current handle atproto-browser.vercel.app/at/anil.reco...
comment in response to
post
it's iterative so far -- what's working really well is building a candidate interface fileset first, then binaries against that (which deliberately fail with a linking error) to check the interfaces expose enough, and then finally actually filling in the module implementations.
comment in response to
post
I wouldn’t depend on this code for anything; it’s strictly throwaway but works surprisingly well for “flat” interfaces which are mostly boilerplate anyway (but not entirely)
comment in response to
post
Your comment lead to a debate among us about what “careful” means — either “have benchmarks!” or “be aware that the parametric memory is unreliable” or generally that evaluating in context is important. Either way, agreed with you — we need to make this sort of qualification more accessible
comment in response to
post
Happy 2.8th birthday!!
comment in response to
post
Just running the Hazel code as OCaml seems easier
comment in response to
post
there really is a plan here that makes this useful! anil.recoil.org/ideas/hazel-...
comment in response to
post
Thank you very much to the pedant who pointed out to me that alumni cannot be "present" by definition. Now REGISTER!
comment in response to
post
Congratulations Bashar!!
comment in response to
post
Well done to Erik Mackie for a fantastic workshop!
comment in response to
post
@dorchard.bsky.social points out this paper from computational biologists on pragmatic guidelines for cross disciplinary work pmc.ncbi.nlm.nih.gov/articles/PMC...
comment in response to
post
@eshuck.bsky.social is the best sheepherder possible for the unruly cambridge lot
comment in response to
post
And @emilyrlines.bsky.social also talked about using lasers in forests but not the burny kind of laser besjournals.onlinelibrary.wiley.com/doi/10.1111/...
comment in response to
post
Round table time to brainstorm what’s next. I’m pushing for speed punting sessions organised by a topic per punt
comment in response to
post
Franks about to do an alpha release of his foundation model in May
comment in response to
post
And last but definitely not least, Frank on his homebrew geospatial foundation model, and fresh from winning the departmental “better future” award! www.cst.cam.ac.uk/announcing-w...
comment in response to
post
Onto SDGs - only 17% are on track esp related to climate and poverty goals
comment in response to
post
Hattie Branson from @florafauna.bsky.social building bridges between AI and urgent conservation needs
comment in response to
post
Rauri on using bioacoustic recorders to model 1000s of hours of wildlife to measure relative abundance across control and treatment sites
comment in response to
post
Orlando is coral chasing to model degraded reefs today and figure out restoration strategies biomin.esc.cam.ac.uk/people/2023-...
comment in response to
post
And then onto Emilio talking about hunting pressure modelling! anil.recoil.org/ideas/mappin...