Can precise time make Gen AI workloads more efficient? | Open Compute Project | Podwise