Cohesix – find out what happened to a local AI job (github.com)

Cohesix has announced the release of version 1.2.0, which enhances edge AI job management by consolidating the entire workflow into a manageable flow. Users can now submit jobs and track their progress with greater accountability, as the system seamlessly reconciles submissions even if a connection drops. This release, which operates on platforms like seL4, QEMU, and Raspberry Pi 4, establishes dedicated GPU paths for both NVIDIA Jetson and Apple Silicon devices, ensuring jobs remain tied to their respective hardware without silent transitions to less optimal environments. This update is significant for the AI/ML community as it emphasizes precise job tracking and recovery processes, mitigating the risks associated with erratic job executions. Cohesix 1.2.0 enables operators to handle model changes purposefully, connecting training, evaluation, and rollback processes while preserving the integrity of rejected candidates. The inclusion of specific download options and clear compatibility requirements reinforces Cohesix's commitment to user-friendly deployment, allowing developers to use existing tools like Python clients and command line interfaces efficiently.