Sr Data Scientist, Compliance

OKX•San Jose, CA

1d•Hybrid

About The Position

This is a role for someone who finds financial crime problems genuinely interesting. At a global crypto exchange, the compliance data landscape is more varied and more analytically demanding than in most financial institutions: on-chain transaction data, cross-jurisdictional exposure, emerging typologies, and a regulatory environment that is still being shaped. The analytical work here is not about maintaining existing models or running standard reports; it is about building the detection capability and analytical intelligence that the compliance function depends on. You will join a team of data scientists, ML engineers, and analytics engineers working closely with compliance specialists across AML, sanctions, KYC/KYB, and transaction monitoring. The role offers a good balance between deep technical work and meaningful engagement with the domain experts who understand what the outputs need to achieve. It suits someone who is comfortable moving between a complex modelling problem and a conversation with a compliance analyst about what a result actually means. AI-assisted analysis is already part of how this team works. We use LLM tooling to accelerate coding and investigation workflows, automated pipelines to replace repetitive analytical tasks, and AI-augmented approaches to surface patterns faster. We are looking for data scientists who already work this way and who can bring that fluency to hard compliance problems, helping the team do better work rather than learning the basics on the job.

Requirements

8+ years in data science, quantitative analytics, or a related field, with experience working on compliance, financial crime, or regulatory analytics problems in financial services, fintech, or a crypto exchange. We welcome candidates across seniority levels; scope and responsibilities will be aligned with your experience.
Solid Python and SQL, with applied experience in machine learning across supervised, unsupervised, and anomaly detection methods, and a working knowledge of how those approaches translate to compliance problem formulations such as typology detection, risk scoring, and threshold calibration.
Experience working with big data platforms such as Spark, Databricks, Hadoop, or MaxCompute, and comfort with both notebook-based exploratory work and structured analytical delivery.
A good understanding of AML, KYC/KYB, sanctions, or transaction monitoring concepts, sufficient to work productively with compliance stakeholders and translate their requirements into analytical workstreams without losing the important nuance.
Hands-on fluency with AI-assisted working practices. You already use LLM coding tools as part of your regular workflow, have experience building or contributing to automated analytical pipelines, and have a practical view on how AI can improve the quality and speed of compliance analytics work. We are looking for people who have put this into practice.
Familiarity with model governance and explainability requirements, and an appreciation for the documentation standards that come with analytical work in a regulated environment.
Good communication skills and the ability to present analytical findings to non-technical audiences including compliance officers, investigators, and legal teams, in a way that is clear, accurate, and actionable.

Nice To Haves

Experience in the crypto ecosystem, familiarity with on-chain data, blockchain analytics, or VASP compliance frameworks is a genuine advantage and will give you useful context for the data and problems you will encounter.

Responsibilities

Develop and validate analytical models and detection strategies for compliance use cases including AML typology detection, customer risk rating, KYC/KYB risk scoring, transaction monitoring calibration, and SAR analytics, working closely with ML engineers on feature design and production implementation.
Translate compliance and regulatory requirements into well-defined analytical problems, working with compliance stakeholders to understand what the models need to detect, where current approaches fall short, and what good performance looks like in practice.
Conduct exploratory analysis on on-chain and off-chain datasets to surface patterns, validate hypotheses, and develop candidate features for detection models, using AI-assisted tooling to accelerate the cycle from hypothesis to validated insight.
Apply AI-assisted coding and automation as standard practice: using LLM tools to write, review, and improve analytical code; building automated workflows that reduce manual query cycles; and contributing to AI-augmented investigation tools that help compliance analysts work through alerts and cases more effectively.
Work with ML engineers and data engineers to ensure analytical work is designed for production, reviewing feature logic, validating data quality assumptions, and providing the domain context that engineers need to build pipelines that behave correctly.
Maintain clear documentation for models and analyses, capturing methodology, data lineage, and performance metrics to the standard required for internal review and regulatory scrutiny.
Support regulatory lookbacks, audit responses, and ad hoc investigations by surfacing relevant data, conducting targeted analysis, and presenting findings to compliance and legal teams in a clear and well-evidenced way.
Stay closely engaged with developments in compliance data science: graph ML, LLM-based investigation tools, anomaly detection, and evolving AML typologies in crypto and fiat environments, with an eye toward what is genuinely production-ready versus what is still experimental.