Add working group meeting notes

docs/working-group/README.md

@@ -10,5 +10,29 @@ Please contribute to our meeting notes by opening a PR.
 2. Work Items
 3. Demos
 
-## November 20th, 2020
-TODO: Notes. See you there!
+# Meeting Notes (12/4/2020)
+TODO: Notes. See you there!	## Attendees
+@ellistarn
+@prateekgogia
+@gjtempleton
+@shreyas87
+
+## Notes:
+-  [Ellis] Shared background
+-  [Guy] Cloudwatch metrics, ECS scaling using cloudwatch metrics for autoscaling.
+-  [Guy] Karpenter supporting generic cloudwatch metrics?
+-  [Guy] Node autoscaling is supported?
+-  [Ellis] Cloud provider like model for cloudwatch, provider model exists in scalable node group side.
+-  [Ellis] Cloudwatch could support Prometheus API?
+-  [Ellis] We can have a direct cloudwatch integration and later refine it?
+-  [Guy] Implementing a generic cloud provider in core in CA.
+-  [Ellis]  Will explore integration with cloudwatch directly, prefered will be coud provider model.
+-  [Guy] Contributions- People in squad will be interested, open to contribute features if it provides value to the team.
+-  [Guy] Scaling on non-pending pods and other resources, people have been asking. Karpenter looks promising for these aspects.
+-  [Ellis] - Long term goal, upstream project as an alternative. As open as possible and vendor neutral.
+-  [Guy] - There is a space for an alternative, given the history CA works around pending pods. Wider adoption possible if mature.
+-  [Ellis] - Landing point will be sig-autoscaling.
+-  [Guy] - CA lacks cron scheduling scaling.
+-  [Ellis] - pending pods are a big requirements.
+-  [Prateek] - introduced the pending pods producer proposal.
+-  [Ellis] - Move time earlier by an hour and change day to Thursday, create a GH issue to get feedback what time works?