Block's Risk Signals and Controls team needs to power risk decisions with trusted signals and effective models by building efficient tools and workflows to accelerate fraud detection and model effectiveness, evaluating thousands of transactions per second in real time.
Requirements
- Reason about complex, distributed systems at high scale
- Proficiency in machine learning techniques, experimental design and data engineering
- Strong programming skills in languages such as Kotlin, Python, TensorFlow, or PyTorch
- Python, Java and Kotlin
- HTTP, JSON, gRPC, Protocol Buffers
- AWS, Databricks, and Kubernetes
- MySQL, DynamoDB, Kafka, Spark
Responsibilities
- Design, build, and maintain data pipelines to ingest third parties and Block data to enhance our feature store and modeling capabilities.
- Design elegant ML pipelines and services, prototype new approaches, and productionize solutions at scale.
- Build tooling to help maintain our features and models, gaining efficiencies in fraud detection.
- Work hand-in-hand with ML Modelers to identify and integrate new data sources, heuristics and models.
- Solve challenging technical problems at scale, collaborating with colleagues located across the globe
- Apply ML and engineering best practices to shape how Block develops, tests, and maintains ML-platform solutions.
- Ensure data quality and completeness through automated validation, monitoring, and alerting.
Other
- 12+ years of experience in software development and demonstrated technical initiative and leadership on previous machine learning projects.
- A proven ability to shape how teams effectively adopt and evolve AI practices; driving adoption of AI-first workflows by coaching leaders, identifying scalable use cases, and embedding quality and accountability in team norms.
- Work autonomously in a fast paced, ambiguous and unpredictable environment
- Natural curiosity & eagerness to learn
- Show the ability to take initiative, lead, and drive creative solutions in cross-team contexts.
