Tomasz Kokoszka

YouTube Ads org is responsible for serving and optimising video ads on YouTube. It is a >$XXB/year business, org size >500 people. I manage a team of 7 engineers. I work closely with Data Science and Data Infra leads to building a reliable end to end data platform with a goal to best capture business impact of the proposed serving system changes. The area is broad, open-ended and ambiguous, it covers data collection, aggregation, orchestration and measurement.

I worked on a traffic estimator - a forecasting system that was directly powering GoogleAds frontend, so advertisers could see estimated ads performance as they create/modify their advertisement campaigns. Before that, I was part of the team that built from scratch AdWords For Video advertiser frontend and API to enable creating video campaigns serving ads on YouTube.

I was a member of a database schema council (supporting the TL, database was storing business entities). I was responsible for reviewing changes to the data schema, and for migrating the database from running on manually-managed machines, to an automated cloud setup. I created staging and a daily refreshed dev instance. In addition to the data part, I was also working on making changes to the customer facing frontend.

I worked on a frontend for an internal build system for CERN internal computation engine. I designed and built the frontend and API for the product web portal (project name: ETICS, technology: Java, Google Web Toolkit).

I worked on a frontend for an internal tool to facilitate designing, deploying and monitoring software that controls the particle accelerator equipment (project name: FESA). I designed and built an API to express system constraints (technology: Java, XML).

Master thesis: “Logical validation of XML documents” (written as part of internship at CERN). Received scholarship for excellent academic performance in 2002 and 2003.