Machine Learning Engineer | Betty CA

Responsibilities

Design, build and ship AI solutions based on LLMs, such as AI agents, retrieval-augmented generation (RAG) and automated analysis and reporting, that teams across the business use every day; Help shape how Betty adopts AI: find high-value use cases, evaluate models and tools, and set up checks that AI outputs are accurate, reliable and cost-effective; Partner with the marketing and user acquisition teams on channel and campaign performance, using metrics such as CAC, ROAS, CPA, first-time depositors (FTDs), conversion rates and predicted LTV; Build marketing models such as predicted LTV for newly acquired players, attribution, marketing mix modeling, incrementality measurement and budget allocation across channels; Work with analysts, marketing, product and business stakeholders to turn business questions into clear requirements, success metrics and use cases; Work with large event-level data in Redshift, and combine and clean data from internal and external sources such as player activity, transactions, payments, CRM and marketing platforms; Build statistical and machine learning models for problems such as player lifetime value, churn and retention, segmentation, forecasting and driver analysis; Form and test hypotheses about player and commercial behavior, and turn the answers into clear recommendations, including how they affect our players; Find opportunities and inefficiencies across player segments, games, providers, payment flows and channels, and run deep-dives that support product launches and marketing campaigns; Support experimentation and A/B testing: define success metrics, analyze results and know when a result is real; Take models from prototype to production and own them end to end, from batch jobs on daily, weekly and monthly schedules to real-time inference where speed matters; Build and run real-time pipelines that stream player events (for example with Kafka or Flink), compute features on live data and serve predictions with low latency, such as real-time churn and retention signals; Set up the MLOps foundations: CI/CD for models, model registry and versioning, a feature store, automated retraining, and monitoring for drift, data quality and model performance; Present results through dashboards and clear recommendations to any audience, up to C-level; Help other teams use data and AI on their own, through self-service analytics and shared environments where they can safely deploy their own AI tools; Work within the rules of a regulated market, including responsible gambling and data privacy requirements

Requirements

3+ years of experience as a Data Scientist/ML Engineer; Strong knowledge of statistics and machine learning methods, such as predictive modeling, classification, clustering, regression, and optimization; Advanced Python with strong software engineering practices (testing, code review, Git), and advanced SQL with Redshift or a comparable cloud data warehouse; Proven experience deploying and running ML models in production, including CI/CD, containers (Docker), model versioning, automated retraining and monitoring for drift and performance; Hands-on experience building solutions with LLMs and generative AI, such as retrieval-augmented generation (RAG), AI agents and tool use, prompt engineering and NLP; Experience working with LLM APIs and frameworks, and with evaluating AI outputs for accuracy, reliability, and cost; Good judgment on where AI adds value and where a simpler method works better, and you keep up with new models and tools; Proven experience applying models to business problems such as forecasting, segmentation, lifetime value, retention, churn, or engagement; Experience with large datasets (100M+ rows and many variables); The ability to work with messy, incomplete, and changing data and still produce reliable results; Clear communication: you can explain complex work to people who don't want the detail, and manage expectations honestly; Comfortable working in a fast-paced environment; Fluent English, written and spoken

Nice to have

Strongly preferred: experience with marketing data and its terminology, such as user acquisition, CAC, ROAS, FTDs, attribution, marketing mix modeling, incrementality, predicted LTV, and campaign and creative performance; Online casino experience from any part of the business, or experience in iGaming, sports betting, gaming, fintech or another high-volume consumer business; Hands-on Tableau or a comparable BI tool; Experience with streaming data and real-time ML: event streaming (Kafka, Flink or similar), feature computation on live data and low-latency model serving through APIs; Hands-on experience with MLOps and orchestration tools such as MLflow, Airflow, Kubernetes, Terraform or a feature store; Experience working with APIs and third-party data sources; Experience running LLM applications in production, including vector databases, agent frameworks, output evaluation and cost monitoring

Benefits

Competitive salary; Performance-based bonus; Career and skills development opportunities; Fun and collaborative team environment; New modern office space