Senior ML/AI Modeler, Risk Automation Machine Learning

Block•Seattle, WA

23h•$160,700 - $283,600•Remote

About The Position

The Risk Automation ML team at Block focuses on automating Risk and Compliance investigations and decision-making through the application of agentic and generative AI technology. The team collaborates globally with partners in Product, Engineering, and Operations to ensure a safe user experience and minimize illicit activity on the platform. They utilize machine learning systems to monitor billions of payment transactions across traditional and blockchain networks, identifying suspicious activity for analyst review. Generative AI is employed to enhance analyst workflows, improve investigative user experience, and accelerate decision-making (Copilots). Additionally, they automate workflows entirely, eliminating manual reviews (Autopilots). This role presents a significant opportunity to optimize Risk Operations at Block at scale. While an individual contributor (IC) role, the senior level involves substantial leadership responsibilities, including owning and driving strategic roadmaps and priorities to completion through cross-functional collaboration.

Requirements

8+ years of Machine Learning modeling experience. Full stack ML experience is strongly preferred.
A Masters or advanced degree in computer science, data science, operations research, applied math, stats, physics, or a related technical field.
3+ yrs experience with AI engineering, Large language models, and a background in traditional NLP techniques is a strong plus for this role.
End to end experience of building and deploying ML/AI to production systems (batch and real time) that are performant at scale.
Experience of independently owning, influencing and driving programs with multiple cross functional stakeholders that have significant business impact.
Have a curious, growth-oriented mindset and the ability to think in first principles to identify creative solutions that demonstrate value.

Responsibilities

Experiment and deploy AI copilot and autopilot systems at scale to improve analyst productivity and/or eliminate manual decision loops altogether.
Own the end to end system including API calls to disparate data sources, advanced prompt tuning, orchestration, metrics and evaluation, productionization and monitoring.
Leverage diverse data sets that include payment transactions, connected users and asset graphs, unstructured text data and user profile information to build transformer based ML models to improve downstream detection tasks.
Work cross functionally with product, platform, engineering and operational stakeholders to deploy production grade systems and monitor and tune ongoing performance.
Use Python ML stack, LLMs, Pytorch, Snowflake, Airflow based tools, data platform and cloud services (both GCP & AWS) to get the job done.
Leverage agentic tools (Claude Code/Codex/Openclaw) to supercharge your research, development, devOps and documentation work as part of your day to day.