As VP of Engineering, I wanted to give our audience a peek “under the hood” of the Engineering group here at Tatari. With over 35 engineers and data scientists, we are working on many impactful technical projects across a number of critical areas in our business. I hope this tour helps our customers and candidates better understand how we’re organized and what we’re working on.
At the highest level, our technical staff at Tatari is organized into three distinct groups: data science, engineering, and product management. Our data science team develops the proprietary measurement techniques and analytics that our customers use to optimize their ad buying strategies. Engineering builds and maintains our client-facing and internal products as well as the required data infrastructure to power them. Product management is the group connecting these two teams to our other stakeholders, including the client services and media buying teams, while discovering and defining the best possible things to build to help our customers grow their businesses.
Broadly speaking, Tatari divides the TV advertising space into buying and measurement.
Our media buying platform engineering team builds the tools and integrations that power Tatari’s ability to buy media across the entire landscape of linear and streaming TV. The challenge for this team runs twofold. First is to abstract away the fractured world of TV advertising into well-designed, streamlined buying tools that supercharge our media buying capabilities at scale. This includes tools we make available to media partners or publishers, helping them to interface with Tatari. Second is to optimize inventory by injecting a dose of data science pedigree in the buying process (e.g. bid optimizations, clearance estimates, etc). This full-stack engineering group works primarily with React, Apollo, GraphQL and Python.
Once all the media orders are final, and the advertisement airs on publishers, such as Roku or ESPN, we then rely on our measurement engineering team to find out how well a campaign performed with audiences. This team is responsible for building the complex data processing and analysis pipelines, distilling the results of both linear and streaming advertising campaigns into intuitive, actionable visualizations and data products. The stack for this team includes React, GraphQL, and Python accessing data sets stored in AWS Aurora, Redshift, and S3. Unlike industry terminology, all technology and data science that relates to planning also falls under measurement.
While the teams mentioned above build our primary customer-facing products, none of this is possible without our dedicated data engineering team. This team is responsible for the entirety of the infrastructure underlying our products, including all data flowing in and out of our systems. Data engineering at Tatari is focused on integrations with first and third party data sets from customers and other data partnerships, providing large scale data processing capabilities in the cloud, developing data schema designs for the efficient storage and retrieval of extremely fine grain time-series data, and the deployment, execution, monitoring, and logging of business critical analytical models and applications in AWS. At Tarari we make extensive use of many open source tools and frameworks, such as Airflow, Vault, and Spark, while also building many custom solutions to fit our specialized needs. Data engineers at Tatari face the unique challenges of operating our expansive data platform for both high-performance, security, and reliability under sustained customer growth.
As Tatari grows, and the TV industry evolves like no other, we are continually encountering new product, data science, and engineering challenges. Innovation is essential. The members of Tatari engineering value humility, technical skill, deep commitment to broad collaboration, and a sense of urgency for delivering value and results. If you’re interested in learning more about our culture, interview process, or open positions please visit our engineering careers page.