Senior /Staff AI Engineer

OKX; Hong Kong; Presencial; HKD 900.000 - 1.500.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Competitive pay

Education subsidy

Team events

Wellness and meals

Healthcare for employees

Descripción de la vacante

OKX is seeking a Senior AI Engineer to design, build, and own identity-risk models. You will fuse verification scores, device/geo data, and user behavior into trust signals, using graph and clustering methods to uncover fraud at scale in real-time.

You will work closely with Product, Risk, and Compliance, shipping models into high-risk production environments and raising the technical bar for the team with advanced ML techniques.

Formación

• 5+ years building risk-control or identity-trust models.
• Graph clustering expertise in fraud or abuse detection.
• Behavioral feature engineering and sequence modeling.
• Proficiency with PyTorch or TensorFlow; production ML experience.
• Strong coding skills in Python and SQL; end-to-end systems.

Responsabilidades

• Design and ship identity trust risk models for real-time decisions.
• Build entity graphs across accounts, devices, faces, and docs to detect fraud rings.
• Develop real-time and offline behavioral feature sequences.
• Convert biometric signals and document checks into reliable risk inputs.
• Collaborate with product, risk, and compliance to productionize models.

Conocimientos

Identity risk modeling

Fraud detection

Graph clustering

Behavioral modeling

Biometric signals

Python

SQL

PyTorch

TensorFlow

Strong coding

Educación

Bachelor's degree in CS/EE or related

Herramientas

Neo4j

GraphX

Spark

Flink

Kafka

Airflow

Snowflake

Descripción del empleo

Who We Are

At OKX, we believe that the future will be reshaped by crypto, and ultimately contribute to every individual's freedom.

OKX is a leading crypto exchange, and the developer of OKX Wallet, giving millions access to crypto trading and decentralized crypto applications (dApps). OKX is also a trusted brand by hundreds of large institutions seeking access to crypto markets. We are safe and reliable, backed by our Proof of Reserves.

Across our multiple offices globally, we are united by our core principles: We Before Me, Do the Right Thing, and Get Things Done. These shared values drive our culture, shape our processes, and foster a friendly, rewarding, and diverse environment for every OK-er.

OKX is part of OKG, a group that brings the value of Blockchain to users around the world, through our leading products OKX, OKX Wallet, OKLink and more.

About the Team

We sit on the front lines of OKX's security and compliance mission. Our charter is to build the flexible, scalable, and intelligent AI systems that detect and prevent identity fraud, account takeover, and large-scale organized abuse - in real time and at the scale of a global financial platform. We own the core identity-risk capabilities that decide, millions of times a day, whether an account can be trusted: identity trust signals, behavioral risk modeling, relationship and cluster intelligence, and face, liveness and document verification. We work shoulder-to-shoulder with Product, Risk, Compliance, and Engineering to turn the hardest adversarial problems in crypto into production-grade defenses that protect good users and the open financial system.

About the Role

We are looking for a Senior AI Engineer to design, build, and own key identity-risk and trust models at OKX. You will build models that combine identity verification results, device and geo data, and user behavior into trust and risk signals, and use graph and clustering methods to uncover organized, industrialized fraud. This is an end-to-end, research-to-production role: you will own algorithms in your domain and ship models into real-money, high-risk, large-scale production environments.

You will be a hands-on technical contributor who works closely with Product, Risk, and Compliance, and who helps raise the technical bar of the team. Your domains span:

• Identity trust risk modeling - fusing verification scores, device, geo, and behavioral data into identity-quality and trust signals.
• Cluster relationship intelligence - graph and clustering methods to identify account farms, mass registration, and coordinated attacks.
• Behavioral risk modeling - behavioral features and sequence models for account takeover and abnormal-session detection.
• Biometric multimodal signals - face embeddings, liveness/anti-spoofing, DeepFake and injection defense, document forgery and AIGC detection, used as inputs to identity trust and the risk layer.
• AI agents - applying agentic AI to scale investigation, detection, and operational defense.

What You'll Be Doing

• Build identity trust risk models: Design and ship coarse-to-fine risk models that fuse face/liveness/document verification scores, device, IP and geo signals, and behavioral data, for high-concurrency real-time decisioning.
• Uncover organized fraud with graph clustering: Build entity graphs across accounts, devices, faces, documents and templates, and apply community detection, clustering, and GNN methods to detect fraud rings and large-scale abuse.
• Develop behavioral features sequence models: Engineer real-time and offline behavioral features, and build sequence models to detect account takeover, bot/script behavior, and abnormal sessions.
• Leverage biometric multimodal signals: Turn biometric features (face embeddings, face quality, liveness and anti-spoofing results) and document-forgery and DeepFake signals into reliable identity-trust inputs for risk decisions, including biometric similarity for cross-account linkage and duplicate-identity detection. Work with the face, liveness, and document teams, or build models yourself.
• Drive execution collaboration: Deliver measurable results while managing ambiguity; partner with backend, product, infrastructure, risk, and compliance teams to integrate models into the production decision chain. Own metric definitions, the balance between false positives and missed fraud, pass rate, monitoring, and drift handling.
• Advance applied R D: Experiment with graph learning, multimodal fusion, and AI agents; stay current with academic and industry advances and translate them into practical defenses.
• Raise the technical bar: Contribute to design reviews, share applied-AI best practices, and mentor junior engineers.

What We Look For

• 5+ years of hands-on experience building risk-control, fraud-detection, or identity-trust models, including significant experience deploying ML systems into production.
• Graph clustering expertise: Experience applying graph mining, community detection, clustering, or GNN methods to fraud or abuse detection.
• Behavioral modeling: Solid experience with behavioral feature engineering and sequence modeling, including real-time feature pipelines.
• Deep AI/ML expertise: Strong proficiency with PyTorch and/or TensorFlow, and solid experience with production ML systems.
• Strong technical coding skills:

Strong coding ability (e.g., Python, SQL) and the system-level thinking to design end-to-end solutions with high availability and low latency.

• Adversarial mindset: Understanding of how fraudsters adapt, and how to evaluate and iterate defenses against evolving attacks.
• Collaboration: Strong communication skills, with experience working across engineering, product, and risk teams.
• Education: Bachelor's degree or higher in Computer Science, Electrical Engineering, or a related field.

Preferred Qualifications

• Experience in identity verification, KYC/KYB, AML, face-payment or payment risk control, or biometric security.
• Hands-on experience with face recognition, liveness/anti-spoofing, OCR, document parsing, fake-ID detection, or media forensics.
• Experience with graph databases or graph computing (e.g., Neo4j, GraphX) and real-time graph features.
• Knowledge of data and AI/ML infrastructure (e.g., Spark, Flink, Kafka, Airflow, Snowflake).
• Familiarity with MLOps best practices: monitoring, retraining, and continuous improvement of production models.
• Experience with LLMs and AI agents.
• Experience working in fast-paced technology companies or security-critical environments.
• Research publications, patents, or open-source contributions in relevant AI fields.

Perks Benefits

• Competitive total compensation package
• L D programs and Education subsidy for employees' growth and development
• Various team building programs and company events
• Wellness and meal allowances
• Comprehensive healthcare schemes for employees and dependants
• More that we love to tell you along the process!

Information collected and processed as part of the recruitment process of any job application you choose to submit is subject to OKX's Candidate Privacy Notice.