Blazer Ops: Automating the development of ML models
2 views
Skip to first unread message
Aaditeshwar Seth
Aug 8, 2026, 1:18:55 AMAug 8
to learning, ac...@googlegroups.com, CoRE stack: NRM, CoRE Stack
Very cool stuff - agents can try out different model variations to keep improving the accuracy, and once the models are trained they can be pushed to github. Remarkably similar to the configurable LULC workflow that Salil is building, and other key elements like STACD to maintain pipeline versioning and cluster job scheduling algorithms to run agentic workflows, which our group is working on, all seem to be key elements of Blazer Ops too. Good validation!
Adi
Co-founder, Gram Vaani; Co-founder, CoRE Stack
Aaditeshwar Seth
Aug 8, 2026, 1:24:08 AMAug 8
to learning, ac...@googlegroups.com, CoRE stack: NRM, CoRE Stack
Correct youtube link: https://www.youtube.com/watch?v=aC642jg2sTA. I don't know why youtube moved on to a prayer for financial breakthrough, following a video on spatial ML workflows!
Reply all
Reply to author
Forward
0 new messages