Responsibilities

• Lead the design, architecture, and delivery of OKX’s in-house Fake ID and document security platform, with a focus on adversarial and batch-based fraud scenarios
• Develop and deploy computer vision models for physical fake ID detection, document tampering identification, and AIGC-generated forgery detection
• Enable cross-user and template-level analysis to identify large-scale and industrialized fraud patterns
• Integrate document intelligence with biometric signals and user risk profiles to support critical risk decisions across key business workflows
• Drive applied research in document security and multimodal AI, and mentor engineers to ensure high-quality, production-grade AI systems

Requirements

• Bachelor’s degree or higher in Computer Science, Electrical Engineering, or a related field, with a strong focus on computer vision, deep learning, and multimodal AI
• 8+ years of hands-on experience building AI/ML systems in areas such as computer vision, multimodal learning, vision-LLM, or generative image/video modeling
• Strong proficiency in deep learning frameworks such as PyTorch, TensorFlow, and OpenCV
• Excellent analytical, problem-solving, and system-level thinking skills

Preferred Qualifications

• Experience working in fast-paced technology companies or startup environments
• Background in identity verification, KYC/KYB, fraud detection, or biometric security
• Prior experience with document OCR, ID parsing, forgery detection, or media forensics