Machine Learning Co-op

Comcast’s Drexel Co-op Program offers an exciting opportunity to gain hands-on experience, build lasting connections, and grow professionally in a dynamic and inclusive environment. This paid, 6-month immersive experience places students at the heart of our business, working alongside talented professionals on meaningful projects that contribute to real outcomes. As a trusted member of the team, you’ll gain exposure to the inner workings of a global media and technology company while developing skills that will serve you well in any career path.

Your experience will include:

• Hands-On Learning & Impactful Work: Tackle real business challenges, collaborate across teams, and contribute ideas that drive results from day one.

• Community, Connection & Giving Back: Build meaningful relationships through social events, peer engagement, and shared experiences. You’ll also have the opportunity to give back through Team UP, Comcast’s volunteer initiative, deepening your connection to both your community and your fellow students.

• Mentorship & Support: Receive guidance from experienced professionals through our dedicated mentorship program, helping you navigate your co-op and beyond.

• Professional Development: Participate in a custom onboarding experience, a curated learning series, and networking events designed to help you build new skills, explore career paths, and gain insights from professionals from across the organization.

At Comcast, we’re committed to investing in the next generation of innovators and leaders. Our Co-op Program is a transformative experience designed to help you grow, connect, and take the next step in your professional journey.

Role Description

Are you passionate about building cutting-edge machine learning models and engineering systems that impact millions of customers globally? Do you thrive on solving complex challenges in content personalization, content monetization, display advertisement, ad-relevancy and scalable infrastructure? Join our highly motivated team at the Arbitration Engine, where we combine advanced machine learning research with world-class engineering to deliver personalized and targeted content on TV platforms like Sky, Xumo, and Comcast. Be part of a team that directly drives revenue through innovative solutions, all while creating exceptional experiences for customers worldwide!

As a Machine Learning Co-op on the Arbitration Engine team, you will work alongside experienced engineers and researchers to develop and optimize machine learning models that power personalized content and targeted advertisements for millions of global customers. You will contribute to building ranking models, designing algorithms to maximize revenue objectives, and improving the relevance of content served to users. Additionally, you will help enhance our scalable and reliable engineering infrastructure, ensuring low-latency delivery of personalized experiences. This role offers the opportunity to gain hands-on experience in applying machine learning to real-world problems, collaborate in a fast-paced environment, and make a tangible impact on the future of content personalization and monetization.

Every day, you will have the opportunity to collaborate closely with Machine Learning researchers to design and implement innovative solutions for personalized content and targeted advertisements. You will work on building robust data pipelines to collect, process, and prepare large-scale datasets for training machine learning models, ensuring data quality and scalability. Additionally, you will contribute to developing pipelines for training machine learning models, optimizing them for performance and efficiency. You will also play a key role in creating frameworks for offline evaluation and validation of new models, enabling the team to assess their effectiveness and refine them before deployment. Furthermore, you will work on setting up and running online controlled experiments (A/B tests) to test the performance of new models in real-world scenarios, measuring their impact on personalization, relevance, and revenue objectives. This role provides a hands-on opportunity to work across the entire machine learning lifecycle, from data preparation to model deployment and evaluation, in a highly collaborative and impactful environment.

What are some interesting problems the student will work on?

In your role, you will collaborate with Machine Learning researchers to explore and experiment with innovative ideas aimed at improving personalization, increasing conversions from targeted advertisements, and enhancing ad relevancy. You will dive deep into understanding the vast datasets available, identifying opportunities to create new metrics and signals that measure the impact of our models and improve their performance. This includes designing and building pipelines to process large-scale datasets, enabling advanced research and experimentation. You will also develop frameworks to evaluate the performance of new models, both offline and in real-world scenarios, ensuring they meet our objectives of scalability, reliability, and user satisfaction.

Some of the exciting problems our team is tackling include optimizing the balance between improving conversions of targeted advertisements and maintaining user engagement, ensuring that revenue growth from monetizable assets does not come at the cost of user happiness. You will work on discovering and incorporating new signals into our data to enhance revenue generation while preserving a seamless and enjoyable user experience. Additionally, you will have the opportunity to contribute to the development of ranking models and algorithms that align with aggregated revenue objectives, helping to shape the future of personalized content and monetization strategies on a global scale.

Where can this student make an impact?

As a Machine Learning Co-op on the Arbitration Engine team, you will have the opportunity to make a significant impact on how personalized content and targeted advertisements are delivered to millions of customers worldwide. Your contributions to building and optimizing machine learning models, designing data pipelines, and developing evaluation frameworks will directly influence the relevance and effectiveness of the content served on platforms like Sky, Xumo, and Comcast. By improving ad conversions, enhancing personalization, and discovering new signals to train advanced models, you will play a key role in driving revenue growth while maintaining user engagement and satisfaction. Your work will not only shape the future of content monetization but also ensure that customers have a seamless and enjoyable experience with the content they consume. Your impact will also include publishing research papers about the ideas implemented to serve customers driving revenue growth while maintaining user engagement.

Job Responsibilities

Responsibilities include but are not limited to:

• Collaborate with Machine Learning researchers to experiment with and implement innovative ideas for improving personalization, ad relevancy, and targeted advertisement conversions.

• Build and maintain data pipelines to process large-scale datasets for training and evaluating machine learning models.

• Develop frameworks for offline evaluation and validation of machine learning models to measure their performance and effectiveness.

• Assist in setting up and running online controlled experiments (A/B tests) to test new models in real-world scenarios and measure their impact.

• Research and identify new metrics and signals to enhance model training and improve revenue generation while maintaining user engagement and satisfaction.

• Contribute to the design and optimization of scalable, reliable, and high-performance engineering infrastructure to support machine learning workflows.

• Other duties and responsibilities as assigned.