Machine Learning Engineer-Xumo

Xumo, a joint venture between Comcast and Charter Communications, was formed to develop and offer a next-generation streaming platform for the entire entertainment industry. The company consists of three primary lines of business: Xumo devices, Xumo Play, and Xumo Enterprise. Powered by Comcast's global entertainment platform, Xumo devices feature a world-class user interface with universal voice search capabilities, making it easy for consumers to find and enjoy their favorite streaming content. Xumo Play is a free ad-supported streaming TV (FAST) service offering hundreds of linear channels and on-demand options. Xumo Enterprise provides tools and services for content creators, distributors, and advertisers to make FAST content more accessible.

Job Summary

XUMO is looking for a highly experienced and motivated ML Engineer to solve business problems by designing and training sophisticated machine learning algorithms and implementing our core recommendation system, which will collaborate with multiple systems to increase product engagement. As the ML Engineer for the XUMO Software Engineering Team, you'll develop complex predictive models using huge amounts of data, create scalable algorithmic solutions to solve business problems, and work with team members to architect and lead the development of the recommendation engine's intelligence. Utilizing tools like Python, BigQuery, and modern ML frameworks, you will build the personalization algorithms that immerse millions of viewers through streaming devices. To succeed in this role, it is necessary to continuously provide clear technical solutions and algorithmic strategies which satisfy the business requirements from the product and operation teams. This job focuses on developing machine learning algorithms for products, training models, and deployment. It involves data pipeline management, technical documentation, and innovation via patents and APIs. The role evaluates ML solutions, conducts case studies, and designs proofs of concept. Team collaboration and junior engineer mentorship are key components.

Job Description

Position Duties

Analyze big data to drive algorithmic insights

• Collaborate with the data team to analyze massive datasets, extracting key user behavioral features and defining quantitative metrics to evaluate model performance.
• Engineer high-quality feature sets and training data schemas optimized specifically for training and serving recommendation algorithms.
• Design and execute rigorous offline model evaluations and online A/B tests, generating deep insights on algorithmic performance against established baselines and benchmarks.

Development of Recommendation System

• Collaborate with product and operations teams to translate business requirements into algorithmic solutions, documenting model architectures and logic to align development tasks.
• Rapidly prototype new recommendation concepts and algorithms, validating mathematical approaches and transitioning experimental models into high-quality, production-ready code.
• Develop the core intelligence of the recommendation system, focusing on implementing, tuning, and scaling advanced machine learning models and the logic for multi-algorithm A/B testing.

Plan and manage recommendation algorithm development

• Collaborate with backend and client engineering teams to integrate machine learning models seamlessly, prioritizing algorithmic enhancements and defining the optimal approaches for model serving and latency reduction.
• Create comprehensive technical documentation detailing algorithmic behavior, model dependencies, and the specifications required for other systems to interact with the recommendation engine.
• Drive strategic decisions regarding model selection, hyperparameter optimization, and feature engineering to solve complex business problems, continuously improving the user experience through highly personalized content delivery.

Qualifications

• 3+ years' experience of using statistical computer languages (Python, R , etc.)
• 3+ years' experience of manipulating big data (Big Query, etc.)
• 3+ years' experience of operating database (MySQL, PostgreSQL, Oracle, MongoDB)
• Experienced and has knowledge of developing variety of machine learning techniques (clustering, decision tree learning, artificial neural networks, etc.) and their real-world advantages/drawbacks.
• Experienced and has knowledge of developing advanced statistical techniques and concepts (regression, properties of distributions, statistical tests, and proper usage, etc.) and experience with applications.
• Strong problem-solving skill with an emphasis on product development
• Excellent written, drawing, and verbal communication skills for coordinating across teams.
• Strong leads with designing, implementation skills to approach the solution for any new features or problem solving
• Self-starter to able to work with minimal supervision for high-quality output
• Aggressive learner for new technologies and techniques.
• Position is office based in Irvine, CA 4 days/week in office & 1 day remote.

Desirable Experiences

• Experience with operating servers on cloud-based environments (Google Cloud, AWS, Azure)
• Experience with data processing framework (Spark)
• Experience with working with Linux-based operation system (CentOS, OSX)
• Experience developing scalable and highly available applications
• Experience with recommendation services

Responsibilities

• Developing and enhancing machine learning algorithms for product and application integration, working with specifications and data pipeline architectures
• Leading the training of machine learning models, overseeing their validation and deployment into production systems
• Creating and maintaining data pipelines, ensuring efficient data ingestion, validation, cleaning, and monitoring processes
• Authoring comprehensive documentation and technical requirements, including evaluation plans, white papers, and reports
• Contributing to the cr