Chris Thrailkill Email & Phone Number

Santa Clara, Ca, Us

Santa Clara, Ca, Us

Los Gatos, Ca, Us

Designing & building scalable data products in support of various operational and analytic initiatives. Highlights include:* Developed a cross-platform utility that enables use of t-digest (https://github.com/tdunning/t-digest) approximate histograms in Spark, Hive, and Druid queries, supporting interoperability between probabilistic sketches created via different versions of the t-digest library. * Contributed to Kayenta (https://github.com/spinnaker/kayenta) OSS Automated Canary Analysis tool, porting the Mann-Whitney U Test algorithm to Scala. * Developed a Spark-based ETL framework - a library of reusable components solving for common ETL development patterns. Used by data engineers & non-data engineers alike, enabling rapid development and self-service data sourcing. * Designed & built a graph (network) based Master Data Management system for maintaining a universal system of record for cloud infrastructure primitives (clusters, instances. etc.) and allowing for federated data governance of corresponding metadata. * Developed the data processing & ML anomaly detection layer of a Security Information and Event Management system. * Developed various time series-based anomaly detection pipelines in support of Payments fraud detection and processing optimization efforts. * Developed, maintained, and supported numerous distributed (Spark, Pig) ETL pipelines, encompassing extremely large data volumes (e.g., high 10’s of billions of rows per hour) and complex processing logic (e.g., high-volume distributed lookups against extremely large, dynamic datasets.)* Prototyped a distributed deep learning training architecture for computer vision models, utilizing Horovod on Spark 3.x and a GPU-enabled cluster. Enabled end-to-end distributed processing from feature engineering through model training.

* Designed, built and managed LikeList data and search platforms. * Developed underlying site database. * Modeled dimensional analytical data mart and built out reports. * Utilized MapReduce on Hadoop to perform ad hoc analysis on voluminous unstructured data. * Created roadmap for scale-out of data systems, accounting for web-scale social growth patterns and addressing transactional, operational, and analytic capability requirements. * Advocated for, planned & managed deployment of data environments in the Cloud, namely Amazon Web Services EC2, EBS, RDS, EMR, and S3 offerings.

Dublin 2, Ie

Data Management and Business Intelligence systems integration, specializing in Customer Data Integration Roles & highlights include:* Enterprise data strategy lead - Was tech lead on effort to define multi-year enterprise data management roadmap and solution architecture, including “to-be” BI and data warehouse architectures and detailed planning of proposed phase 1 master data management (MDM) solution implementation. * Data architect/engineering lead - Led multiple phases of an enterprise data integration project. Tasked with scoping, planning, architecting, and delivering a consolidated customer data mart, integrated across all lines of business. Role included management of offshore resources. * Development lead and performance test architect - Developed instrumentation to assess performance and scalability of real-time ETL flows deployed as SOAP services. Analyzed and managed deployment of performance code fixes. * Data engineer - Designed and built operational data processing solutions for ad hoc, rapidly deployable analytics initiatives. * Developer, systems admin, and DBA - Supported delivery of various business intelligence proof-of-concept initiatives. Performed tasks such as deployment and administration of servers, storage, and databases; database configuration and tuning; platform deployment and administration; dimensional data modeling, ETL and report development.