Forward-deployed Data Scientist II

Braze Braze · Enterprise · London, United Kingdom · Customer Experience

The Forward-Deployed Data Scientist II at Braze will design and build end-to-end machine learning solutions for customer personalization, owning the full ML pipeline from data transformation to model activation. This role involves driving customer success, extending product capabilities, and contributing to the strategy and roadmap of Braze's AI products, particularly its reinforcement learning algorithms.

What you'd actually do

  1. Design ML use cases from the ground up — scoping solutions that optimize for real business value, accounting for the complexity of modern marketing journeys, and proactively identifying risks to set each engagement up for success
  2. Build and own the full ML pipeline — taking customers' raw data through transformation, model training, and activation, so that model decisions are delivered to personalize experiences for millions of end users
  3. Drive customer success by providing ongoing technical guidance that ensures data science performance, successful adoption and measurable outcomes
  4. Extend product capabilities by developing features and tools that support the broader AI deployment team and scale what's possible across engagements
  5. Partner with the Braze Product team to refine and advance Braze's reinforcement learning algorithms, pushing the self-learning capabilities of the platform forward

Skills

Required

  • Python (Pandas)
  • TensorFlow
  • Keras
  • scikit-learn
  • CatBoost
  • XGBoost
  • SQL
  • machine learning pipelines
  • model deployment
  • well-structured, modular, documented code
  • strong development practices (Git, CI/CD, testing frameworks, type-hinting, code reviews)
  • scalable, maintainable solutions
  • working directly with clients
  • cross-functional teams
  • aligning stakeholders
  • translating technical concepts into clear business value
  • identify opportunities and risks early
  • troubleshoot obstacles
  • drive creative solutions
  • stay current with industry trends
  • explore new tools/technologies
  • explain complex technical ideas persuasively to both technical and non-technical audiences

Nice to have

  • DevOps tools (Airflow, Kubernetes, Terraform, GCP)
  • data integration/ETL
  • pipeline optimization
  • reinforcement learning algorithms

What the JD emphasized

  • production environments
  • reinforcement learning algorithms

Other signals

  • design and build end-to-end machine learning solutions
  • power 1-to-1 personalization
  • optimize for real business value
  • model training, and activation
  • personalize experiences for millions of end users
  • extend product capabilities
  • reinforcement learning algorithms
  • self-learning capabilities of the platform
  • BrazeAI product strategy and roadmap