Technical Project Manager, SEON

• We are currently looking for a rare combination of research brilliance and pragmatic engineering, someone who can transform cutting-edge computer vision and NLP research into high-assurance AI systems for regulatory compliance and identity verification
• Deep expertise in computer vision, particularly document understanding, OCR, and information extraction
• Strong understanding of modern NLP techniques, especially those applicable to entity recognition, name matching, and contextual understanding
• Experience designing and implementing high-assurance AI systems with appropriate verification, validation, and explainability
• Demonstrated ability to translate research into working prototypes and production systems
• Expertise in LLM behavior engineering for large language models and multimodal AI systems
• Demonstrated ability to design and implement effective evaluation frameworks for AI models
• Experience with systematically improving AI model performance through iterative testing and refinement
• Applied Research experience in Computer Science, Electrical Engineering, or related field
• Publication record in relevant research areas is a plus
• Entrepreneurial mindset with the ability to operate independently and build from first principles
• Excellent communication skills with ability to explain complex technical concepts to both technical and non-technical audiences

Desirable

• As the founding member of SEON's AI Innovation team, you'll drive a research agenda that reimagines how document authentication, identity verification, and AML screening work in production environments
• Your solutions will need to meet the exacting standards of financial regulators while delivering tangible advantages in accuracy, explainability, and cost-effectiveness over legacy systems
• This is an opportunity to build something from first principles
• You'll work directly with C-level executives to shape our AI strategy, develop production-ready prototypes, and establish our technical approach to high-stakes decision systems in a regulated environment
• Success means creating solutions that can be validated with mathematical rigor, deployed into production environments, and serve as the foundation for SEON's next generation of AI-powered fraud prevention products
• Design and implement advanced architectures for document understanding that can extract, normalize, and verify complex identity data from diverse document types
• Create novel solutions for AML name screening that incorporate context-aware matching, transliteration handling, and explainable decision frameworks
• Develop evaluation methodologies that rigorously benchmark new approaches against both academic standards and real-world performance metrics
• Design sophisticated AI interaction architectures that optimize model performance across document analysis and fraud prevention systems
• Build systematic frameworks for collecting and analyzing model performance data to guide continuous improvement cycles
• Architect end-to-end systems for document analysis that combine computer vision, OCR, and structured information extraction with robust verification logic
• Build high-assurance systems with appropriate guardrails, confidence scoring, and formal verification methods
• Develop APIs and integration patterns that allow seamless incorporation of AI capabilities into SEON's broader platform
• Create comprehensive evaluation suites for measuring model accuracy, fairness, and reliability against established benchmarks
• Establish methodologies for training, fine-tuning, and continuously improving AI models with domain-specific data
• Evaluate emerging research, tools, and platforms from major cloud providers and research institutions
• Develop product-focused neural reasoning frameworks that balance performance with explainability requirements
• Create innovative evaluation methodologies that validate AI systems against both technical and business metrics
• Partner with product management to translate research innovations into market-ready capabilities
• Collaborate with compliance and legal teams to ensure AI systems meet regulatory requirements for explainability and audibility
• Build relationships with academic researchers and industry partners to accelerate innovation
• Mentor and guide engineers as the Austin AI Innovation team grows