Christopher Zimmerman Email & Phone Number

San Francisco, Ca, Us

San Francisco Bay Area

Lead Quantcast Engineering and Corporate Technology. We develop and operate a bespoke technology stack that leverages proprietary data, machine learning, and artificial intelligence (AI) at scale, to autonomously make millions of ad buying decisions per second. In a moment in time, these systems use a multi-goal predictive control framework and continuous learning to calculate the optimal value of an ad impression balanced across a set of customer goals. The net result is a platform that individually optimizes every ad campaign continuously to deliver outstanding performance.—Delivered self serve platform.Reduced cost of service by tens of millions of dollars, while improving performance and addressable ad inventory.Grew London based engineering 4x and maintained attrition below the industry average.Rebuilt Corporate Technology organization from the ground up and introduced company wide information security function.Partnered with the CLO to fully comply with GDPR, CCPA, and other privacy regulations, combining both bespoke technology solutions with regulatory compliance processes.Refined and implemented a performance management system that is evidence based, with checks and balances to drive fairness and objectivity across all of engineering.Developed and implemented a career pathing system for engineering managers and individuals contributors, that is easy to reason about; clearly articulating the 5 most important operating characteristics necessary to move to the next level. Worked with Human Resources to generalize the manager leveling criteria and implement it company wide.Navigated organization through a global pandemic and associated business impacts.

San Francisco, California

Led the Infrastructure Platform. These platform teams develop, maintain, and operate the petascale distributed storage and processing systems that support the business, along with a set of common tools and services that provide the foundation on which Quantcast operates. The platform owns the Quantcast FIlesystem, a space efficient highly performant distributed file system (https://github.com/quantcast/qfs). A proprietary map/reduce implementation that is four times faster than leading open source alternatives, which includes efficient distributed resource allocation, a job auto tuning system, data catalog, SQL-on-Map-Reduce service, and self-service data pipeline management.In addition to distributed computation and storage, the platform owns realtime edge counter streaming and aggregation, global event collection, and petascale data transport. In support of these responsibilities, the platform owns Quantcast's hardware and cloud infrastructure, including a highly available dedicated network backbone, that connects all of Quantcast's points of presence globally. Driving productivity and operational excellence across engineering is a core responsibility of the platform. In this area, teams develop and operate Quantcast's build, deployment, universal ticketing, and a nascent services platform.Of note, the platform ingests 1 PiB of data per day, houses 12 PiB worth of live compressed data, and processes close to an exbibyte per month.

San Francisco, California

Led the Big Data Platform group, which consisted of multiple teams that developed and operated the petascale distributed systems that support the business. This included realtime edge counter streaming and aggregation, data pipeline scheduling, efficient distributed resource allocation, a data catalog, the proprietary map/reduce stack, a SQL to map/reduce system and the Quantcast File System.The group successfully migrated these systems to Amazon Web Services with zero customer impact. This included the migration of a complicated proprietary software stack spanning thousands of machines, but also thousands of interconnected data pipelines with hundreds of out of band dependencies.

San Francisco, California

Led the Platform Operations SRE team responsible for the operations of Quantcast's large scale distributed systems, including event collection, realtime bidding, and petascale data processing.Reduced the number of on-call incidents per week from an average of 30 to less than 1.

San Francisco, California

Led the Cluster Operations SRE team which owned and operated the compute cluster and numerous instances of the Quantcast File System. Drove hours of impacted operator time from 300 to less than an hour per week. At the same time improving cluster availability from 95 to 99.5% per week. During this period data processed per day grew from 0.4 PiB per day to 15 PiB per day.

San Francisco, California

Contributed to the development and operation of Quantcast's proprietary map/reduce cluster.

San Francisco

As part of a major effort to move TypePad to a service oriented architecture; developed a horizontally scalable in memory service for reading and writing text blobs, organized in an arbitrary hierarchy. The system is used to power blog comment threads.Developed a system that made it trivial for engineers to insert arbitrary tagged timers into the TypePad stack. These metrics are then aggregated and organized into a chronological hierarchy of tags, that are then visualized to identify performance bottlenecks and regressions.

Santa Clara

Worked on grid computing, specifically Hadoop on Demand, a system for dynamically provisioning Hadoop map/reduce and distributed file system clusters, for use by researchers and developers at Yahoo!Developed an XML-RPC based relay messaging system. The system allows for both asynchronous and synchronous messaging via XML-RPC call backs.Led an effort to develop a methodology for testing the Yahoo! search engine and the corresponding framework. This included a system for teeing live load, replaying load, a database store and a web UI for managing, executing, and analyzing test runs. This work directly contributed to the successful launch of the Yahoo! Search Technology, replacing Yahoo's reliance on Google.

Palo Alto

Worked with the All The Web search marketing services group to integrate their systems with AltaVista's paid inclusion efforts.

Palo Alto

Developed a system for automated building and maintenance of small search indexes. Specifically, a news and freshness index. The system accepts feeds from the central crawlers, filters them, indexes them, and maintains statistics on how frequently content changes via a finger printing system. All this information is then fed back to the central crawlers and global algorithms cluster. It is built atop of several AltaVista technologies and backed by a relational database.Led the move to commodity based hardware, away from the DEC Alpha. Performed hardware reliability and performance evaluations of dozens of white box configuration. The work required porting some key AltaVista technologies to Linux.