
An early-stage experiment. Originally a digital literacy nonprofit; currently being repurposed as an RLHF and human-feedback platform. Underbaked and exploratory.
Project Tags
Buddhimanbaccha started in 2019 as a small digital literacy effort — training a cohort of underprivileged students in basic ITeS skills for back-office work. The pilot ran its course and the project sat dormant.
The infrastructure and the cohort relationships didn't disappear, though. With the rise of human feedback as a core input to model training, the same network of trained annotators turned out to be unexpectedly relevant.
Currently being rebuilt as a small RLHF and human-feedback platform. Early days — task design, annotator workflows, and quality controls are all still being figured out. No claims about scale or output yet.
Sharing it here mostly as a note that it exists and is moving. More to come if and when it earns a fuller writeup.