# Christos Chrysafis > Apple Location: San Francisco Bay Area, United States Profile: https://flows.cv/christoschrysafis * Cloud services, databases, distributed systems, C++, C, Linux, Embedded systems, Java, Linux, Kubernetes. * SAAS (Software as a service). * Service evolution ## Work Experience ### Staff Software Engineer @ Apple Jan 2016 – Present | Cupertino Large scale databases, server to server services, user private+public data. Large scale personalized search/indexing. Security, deployment and build systems. Blob storage. State on the cloud for all devices. Large redesigns, deployments, java backends. [A Multi-Tenant Structured Datastore] https://www.foundationdb.org/files/record-layer-paper.pdf ### Staff Software Engineer @ LinkedIn Jan 2015 – Jan 2016 | Sunnyvale, CA Member data processing, data standardization pipelines, big data mining, real-time updates. Rearchitecture of legacy systems, unification of real time and offline systems. Database vs hadoop-mapreduce. apache kafka, apache pig, apache samza, apache lucene, apache flink, apache spark, avro schemas, apache hadoop, hdfs, data streaming/pipelines; rest apis, svn; ### Staff Software Engineer @ Google Jan 2010 – Jan 2015 | Mountain View CA Knowledge Graph (KG), and Knowledge Panels (KP). Answer queries like [academy awards] or [Tom Hanks awards], [godfather characters]. Responsible project delivery, query triggering, UI, and data for several KP verticals (vertical=Effort to bring query answers to the user in a specific area using KG). Determine user intent for each query, using query annotation and grammar. Retrieve related facts using the Knowledge Graph (KG) with a custom realtime database access language. Display data to the user information, UI configuration (html). Page annotation with mids from freebase and Google's Knowledge Graph. Annotate the world, annotate all existing pages in the world within a period of a day. Improve precision, introduce new quality metrics. Annotate all incoming query traffic. Autocomplete. Responsible for URL autocompletion for google.com as well as the Chrome browser. Autocompletion is based on URL popularity as well as personal history. Query Refinements. Using logs history to decide on refined queries for a search, for [Barack Obama] see queries at the end of the page. Significant improvement in user engagement, and click-through rate Readability, C++, Python, Javascript. ### Staff Software Engineer @ NVIDIA Jan 2006 – Jan 2010 | Santa Clara, CA Early days GPU computing, computer vision, pioneered CUDA development, ### Staff Software Engineer @ ESS Technology, Inc. Jan 2000 – Jan 2006 | Fremont/Sunnyvale ESS Technology acquired Divio Inc. The original Google Glass, failed 10 years before Google Glass's failure. http://www.amazon.com/Deja-View-CamWear-Model-100/dp/B00063E872 Video streaming. ### Staff Software Engineer @ Hewlett-Packard Jan 1997 – Jan 2000 | Palo Alto CA Image compression. Less space more pixels. Color laser printer algorithms. ### Summer Intern @ Conexant/Rockwell Jan 1997 – Jan 1997 | Newport Beach, CA ITU-T30e ### Research Intern @ Xerox Jan 1996 – Jan 1996 | Webster, NY XRCW (Xerox Research Center Webster) ## Education ### Doctor of Philosophy (PhD) in Electrical Engineering, Minor Math University of Southern California ## Contact & Social - LinkedIn: https://linkedin.com/in/christos-chrysafis-4a7a7216 --- Source: https://flows.cv/christoschrysafis JSON Resume: https://flows.cv/christoschrysafis/resume.json Last updated: 2026-04-12