I am Terra Field, a Staff Platform Engineer at Honeycomb!
Previously I was a Senior Software Engineer at Netflix.
Prior to that, I was a Senior Systems Engineer working on the Data Platform at Blizzard Entertainment.
My work:
My current title is Staff Platform Engineer, though I still think of myself primarily as a Systems Engineer. I glue stuff together to make computers do stuff, mostly with Hadoop, Kafka, & Linux. When everything falls apart, I try to figure out why it broke, fix it, and try to make sure it breaks differently in the future.
What I look for in a team: I am a very social creature. I prefer teams that are collaborative and high communication (even when doing solo projects). Especially when doing remote work, I feel more connected to a team that has a high amount of chat activity - even if a lot of it is just banter instead of specifically work or project related.
Roles at Honeycomb (May 2022 - Present)
I recently joined Honeycomb and will be working on scaling and improving their Apache Kafka infrastructure as well as contributing to the overall platform.
Senior Platform Engineer
May 2022 – Present
Roles at Netflix (February 2019 - November 2021)
Senior Software Engineer, Operating System/Compute Team
April 2021– November 2021
Created a service to emit metrics and events when the oomkiller is activated on any of our EC2 instances to help right-size instance types and reduce unexpected service restarts
Part of a four person rotation for on-call support for the base Linux operating system on all instances and containers
Leadership, Trans* Employee Resource Group (ERG)
February 2019 - November 2021
Leader of the Education and Talent Acquisition Pods
Worked to advocate for and represent transgender and gender non-conforming individuals in the workplace, increasing awareness, visibility and education about the trans experience
Worked with the I&D department to develop and deliver training on how to be a better colleague to trans and gender non-conforming people
Consulted with content creators on how to improve the representation of trans people both in front of and behind the camera
Senior DevOps Engineer, Personalization Infrastructure
February 2019 - April 2021
Supported a team of 60 developers and researchers in training and deploying machine learning models for personalized recommendations (row selection, box art selection, billboard selection, etc.)
Deployed and maintained Apache Spark clusters consisting of Mesos agents running on tens of thousands of EC2 instances
Implemented lineage and cost tracking, allowing us to track the cost effectiveness and data fidelity of our workflows
Helped develop, test, and deploy a method of sharding Apache Mesos to overcome scaling limitations encountered at 12,000+ node counts
Roles at Blizzard Entertainment (January 2015 - January 2019)
Senior Systems Engineer, Big Data
February 2018 - January 2019
Deployed Kafka clusters on Kubernetes using StatefulSets and Metacontroller and securely mirrored production data to them to support moving portions of our Global Data Platform into the public cloud.
Eliminated licensing costs, increased performance and reliability, and enabled new features by planning and executing a zero-downtime migration of our production Kafka clusters that handle 20+ billion events per day from a commercial Kafka distribution to Apache Kafka
Reduced costs and increased operational insight by replacing commercially licensed Kafka monitoring and deployment with a combination of Puppet, Telegraf, Jolokia, and Grafana
Debugged open source and in-house applications in Scala, Java, and Python to assist developers in increasing the performance of their applications and to optimize cluster usage
UNIX Systems Engineer
January 2015 - February 2018
Planned and executed a zero-downtime migration of aging 3 petabyte Hadoop-based data warehouse to newer hardware utilizing Kafka to dual write to both clusters and working directly with our analysts and data scientists to migrate jobs. Increased compute and memory capacity by 400%, allowing us to utilize in-memory computing technologies such as Apache Spark and Impala.
Reduced execution time for business reports from hours to minutes by working with business intelligence teams to migrate jobs from Streaming MapReduce to Hive, Spark, and Impala SQL.
Increased message capacity and saved trillions of application events by migrating existing global fan-in RabbitMQ-based data pipeline to use Kafka, greatly increasing customer confidence in our data pipeline
Significantly contributed to the design, testing, and deployment of a modernized ETL pipeline that made data more compact, performant, and accessible by converting it to Parquet and inserting into Hive
Integral in the design, testing, operation, and deployment of Blizzard’s Telemetry pipeline for monitoring service health for all games and services
Designed and deployed a globally distributed ZooKeeper architecture to support service discovery for Overwatch