Senior Machine Learning Engineer - Ads

About the Role

As a Senior Machine Learning Engineer, you will play a crucial role in the development and implementation of cutting-edge artificial intelligence products. Your responsibilities will involve designing and constructing sophisticated machine learning models, as well as refining and updating existing systems. To thrive in this position, you must possess exceptional skills in statistics and programming, as well as a deep understanding of data science and software engineering principles.

This is a hybrid role that can be based in either Hyderabad or Bengaluru, with some expectations to come into the office. You will report to a Machine Learning Engineer.

The Data & Science Org

At Deliveroo, we have a world-class data science organisation with a mission to enable the highest quality human and machine decision-making. We work throughout the company - in product, business and platform teams to answer some of the most interesting questions out there. Data Scientists and ML Engineers at Deliveroo report to our data science management team, and we have a strong, active data science community with guest lecturers, a robust technical review process, a career progression framework, and plenty of opportunities to learn new things.

The Ads ML Team

We are part of the cross-functional Ads Tech at Deliveroo, a group of 20+ passionate engineers, scientists, and machine learning engineers who focus on building our internal Ad Platform, where our Partners can advertise their restaurants.

The Ads ML Team is responsible for recommending sponsored content on Deliveroo. The team strives to push the boundary of what's possible in this space, working on Click-Through-Rate estimation, Automated Bidding, Budget Recommendation, and Dynamic Ads Load. Apart from excelling in complex prompt engineering, there will be a focus on developing in-house models and auction simulators, deploying models in production, and improving ML model monitoring and alerting. We evaluate the performance of all our decision-making machines via robust experimentation powered by our world-class experimentation platform.

What You'll Do

• Work in a cross-functional team alongside engineers, data scientists specialised in analytics and inference, and product managers to develop systems that make automated decisions at a massive scale.
• Use frameworks such as Tensorflow to build new and improve existing machine learning models for sponsored content recommendation, automated bidding, budget recommendation, and dynamic ad load.
• Automate training and inference pipeline using job orchestrating frameworks such as Argo Workflow.
• Mentor and coach team members on ML engineering best practices.

Requirements

• A master's degree in Computer Science, Mathematics, or a related quantitative discipline.
• 5+ years' experience deploying machine learning models at scale in production environments.
• Proficiency in writing production-quality Python code.
• Familiarity with Python data science and machine learning libraries, including scikit-learn, TensorFlow, Keras, pandas, numpy, and XGBoost.
• Thorough understanding of best practices in MLOps.
• Sound knowledge of the machine learning application development life cycle, encompassing CI/CD, version control (git), testing frameworks, MLOps, agile methodologies, monitoring, and alerting.
• Hands-on experience in operationalising ML models and constructing ML pipelines.
• Proficient in exploratory data analysis, model/algorithm prototyping and selection, and model pipeline design and development.
• Practical experience with AWS or a comparable cloud service provider.
• Comfortable working with Docker and containerised applications.
• Familiarity with Git, GitOps practices, Argo Workflow, etc.
• Familiarity with CI/CD tools such as Jenkins, CircleCI, GitHub Actions, etc.
• Strong collaborative skills, able to work alongside scientists, engineers, and non-technical stakeholders.
• A bias for simplicity and impact.
• Experience developing Click-Through-Rate estimation and automated bidding models.