Update README.md

Signed-off-by: Claudia Misale <[email protected]>

Author: cmisale

README.md

@@ -1,5 +1,7 @@
 # AI Training Autopilot
 
+[[README refresh ongoing]]
+
 Autopilot is a Kubernetes-native daemon that continuously monitors and evaluates GPUs, network and storage health, designed to detect and report infrastructure-level issues during the lifetime of AI workloads. It is an open-source project developed by IBM Research.
 
 In AI training jobs, which may run for weeks or months, anomalies in the GPUs and network can happen anytime and often go undetected. In this case, performance degrades suddenly and a deep diagnostic is needed to identify the root cause, delaying or deleting the current job. Similarly, hardware anomalies can greatly disrupt the throughput and latency of an AI inference server.