Data Scientist I

About GumGum

GumGum is The Mindset Company™ transforming advertising. We’re an advertising technology company delivering results by matching brands with people in the right mindset in the moments that matter. Our platform is powered by the Mindset Graph™, our AI-driven data engine that processes billions of real-time contextual, creative, environmental, and historical signals to match every ad with the most receptive audience. The result is advertising that drives meaningful outcomes for advertisers and publishers, and is more relevant for consumers.

We were founded in 2008 and are headquartered in Santa Monica, California, operating in over 19 markets across North America, Europe, Japan, and Australia.

Our principles guide our work every day and are as follows:

• Customer-Obsessed: We’re focused on advertising solutions that solve needs and drive success for clients and partners.
• Make it Happen: We have a bias for action and take ownership to deliver results.
• Always Innovate: We push boundaries with creativity and technology.
• Foster Belonging: We ensure colleagues feel included, supported, and empowered to thrive.

About the Role

The Data Scientist I supports statistical analyses of large datasets, the development and deployment of Machine Learning (ML), multimodal Deep Learning (DL), and Artificial Intelligence (AI) solutions that improve the relevance and value of ads across our Ad Exchange, Contextual Platform, and Attention Measurement Platform. This role focuses on applying strong analytical foundations, building and evaluating ML and DL models, leveraging general AI concepts and techniques, while also partnering closely with the Engineering, Product, and Data Science team to improve ad-serving performance, operational decision-making and lead development of new AI products that best serve the business.

The ideal candidate is curious, data-driven, and eager to develop a strong understanding of the advertising domain and the systems that power large-scale decision-making. You will work with large datasets, contribute to production AI systems, and gain hands-on experience across the ML lifecycle — from exploration to monitoring.

Note: GumGum fosters a flexible work environment, offering GumGummers the ability to work either in-office or remotely/from home. For this position, in-person/office collaboration is required 2 days per week, supporting a balanced approach to flexibility and team engagement.

What You'll Achieve

• Support the translation of business and product requirements into data-driven analyses and ML solutions
• Partner with Engineering team members and senior Data Scientists to develop, test, and deploy ML and DL models
• Conduct exploratory data analysis to inform feature development and modeling approaches
• Build, run, and maintain regular pipelines to analyze production data, generate KPIs, and prepare automatic retraining of existing models
• Query, clean, and structure large datasets using SQL, Spark, and cloud data platforms
• Train, evaluate, and iterate on traditional ML models and multimodal deep learning models under guidance from senior team members
• Design and maintain Looker dashboards and other Business Intelligence (BI) tools to track Key Performance Indicators (KPI) for key stakeholders
• Develop and deploy agentic pipelines and other LLM-powered applications, including prompt engineering, tool use, and evaluation of model outputs
• Contribute to existing Machine Learning Engineering (MLE) workflows for model training, deployment, and monitoring
• Document analyses, models, and broader learning to support knowledge sharing across the team and non-technical audiences
• Continuously expand on statistical and AI foundations while learning new AI/ML techniques, tools, and advertising-domain concepts

Skills You'll Bring

• Bachelor’s degree in a quantitative field (e.g., Statistics, CS, Math, Physics, or Economics).
• 1–2+ years in a data-driven role such as Analytics, Data Science, or ML Engineering.
• Proficiency in Python and experience applying ML/DL methods using libraries like scikit-learn, PyTorch, HuggingFace, or OpenCV.
• Dependable SQL skills and experience designing pipelines or DAGs using tools like Airflow or Astronomer.
• Exposure to cloud environments (AWS/GCP, Databricks, Snowflake) and large-scale query tools like Spark or Snowpark.
• Strong grasp of A/B testing, experimental design, and statistical concepts (regression, classification, optimization).
• Experience with LLM prompting and interest in frameworks like RAG, LangChain, or agentic systems.
• Ability to design dashboards for diverse audiences and collaborate effectively with Product and Engineering teams.