Who We Are Yieldmo is an advertising technology company that operates a smart exchange which differentiates and enhances the value of ad inventory for buyers and sellers. As a leader in contextual analytics, real time technology, and digital formats, we create, measure, model, and optimize campaigns for unmatched scale and performance. By understanding how each unique impression behaves and looking for patterns and performance in real time, we can drive real performance gains without relying on audience data. What We Need We're looking for an experienced architect and engineering leader to spearhead our machine learning operations, with a focus on optimizing the deployment and management of ML models. You will design and implement systems to support our data science initiatives in a low latency high volume (500B - 1T transactions a day) production environment. You are experienced with the lifecycle of machine learning, data science, pipeline designs and have a deep understanding of how they can be deployed in a cloud based infrastructure. Ideally you would have worked as a data scientist for some years and decided your passion lies in the ML ops / data engineering side of data science. We currently have multiple implementations in place that we'd like to unify into a common framework to ensure security, support reusability, observability and to expedite the ability to make changes. You should be capable of implementation in addition to design. You will partner with our data scientists to set best practices, automate data science approaches, architect solutions, deploy and monitor production workloads. Responsibilities
- Architect, deploy, and manage a software framework to securely handle ML models in production-grade environments
- Facilitate seamless integration of ML models into operational pipelines between engineering and data science
- Design, implement and maintain scalable, efficient ML pipelines on AWS and GCP
- Design, implement, and maintain CI/CD pipelines to automate the continuous integration and delivery of ML models
- Monitor performance of ML models and implement improvements
- Ensure high availability and fault tolerance of the ML infrastructure
- Optimize the performance and cost-efficiency of ML systems
- Maintain and champion ML ops best practices
- Manage team of data / ML ops engineers
- Recruit, mentor and grow machine learning engineering team
- Contribute to and evolve our current data science initiatives around (cookieless targeting, format optimization, creative optimization, throttling, margin management)
Requirements
- 10+ years of experience with AWS or GCP.
- 10+ years of experience in software development with Python focused on ETL, ETL and pipeline work
- 10+ years of experience working with large datasets (working with raw logs in the order of 200 - 250TB a day. Comfortable working with databases with tables ranging from 100's of billion to trillions of rows.)
- 5+ years of experience working with data science technologies such as (Python, R, SQL, TensorFlow, PyTorch, scikit-learn, and Spark) as a data scientist
- 5+ years with predictive analytics and developing optimization algorithms
- 5+ years doing machine learning with high dimensionality
- Experience working with Airflow
- Expert level understanding of SQL in Snowflake and MySQL environments
- Experience with Hadoop, Apache Hive, Apache Kafka, AWS (SageMaker), Google Cloud (BigQuery, Vertex AI)
- MS or equivalent combination of education and experience
- Ad tech experience (SSPs, DSPs, Analytics, DMPs, CDPs)
Nice to Haves
- Tensorflow, Scikit, Google Vertex AI experience
- Exposure to A/B testing
- Comfortable reading code in Java
Perks
- Home office setup stipend
- 1 Mental Escape (ME) days each month
- Learning stipend and professional development opportunities
- Work life balance, flexible PTO and competitive compensation packages
US Jobs: The base salary range for this role is: $250,000-$300,000 per year. The range listed is just one component of Yieldmo's total compensation package for employees. Individual compensation decisions are based on a number of factors, including experience, level, skillset, and balancing internal equity relative to peers at the company. We recognize that the person we hire may be less experienced (or more senior) than this job description as posted. In these situations, the updated salary range will be communicated with you as a candidate. For all other countries, we have competitive pay bands based on market standards.