From August of 2014, I began my transformation into a proper SRE type. Working as a two-man team with Michael D'Auria, I did the DevOps for Crowdtap for right about a year.
We delivered (with CircleCI(?)), monitored (with Nagios and later consul health checks), and saw to the care and feeding of a bunch of Ruby on Rails microservices sharing data in RDS and self-hosted Mongo via a RabbitMQ (and later Kafka) event sourcing dataplane, in ASGs on AWS, with baked images configured by Chef; with Puppet for the dev environment management (remember Boxen?) initially. I stuck around this job just long enough to make the jump to Docker on the dev side and see the NYC version of Bryan Cantrill's first talk in The Container Trilogy as part of a Joyent ~roadshow.
We delivered (with CircleCI(?)), monitored (with Nagios and later consul health checks), and saw to the care and feeding of a bunch of Ruby on Rails microservices sharing data in RDS and self-hosted Mongo via a RabbitMQ (and later Kafka) event sourcing dataplane, in ASGs on AWS, with baked images configured by Chef; with Puppet for the dev environment management (remember Boxen?) initially. I stuck around this job just long enough to make the jump to Docker on the dev side and see the NYC version of Bryan Cantrill's first talk in The Container Trilogy as part of a Joyent ~roadshow.
