Not really suitable for the common scalability issues startups deal with today. Like working in multiple Amazon regions or supporting difference sets of servers.
you're well aware that the for example the couchbase cross-datacenter-replication has it's own share of problems such as "what happens when the same dataset gets modified in both clusters?". IIRC it just drops the older change and keeps the newer version. That may or may not be a problem to you, but for others that might just be the nail in the coffin. Every datastore out there has different trade-off that are acceptable for different use-cases. And postgres has it's own share of tradeoffs, but it works quite nicely for a lot of use-cases.
Every notable PostgreSQL deployment has had to 'roll their own'.