Director, AI Engineering

About Us: Structure Therapeutics develops life‐changing medicines for patients using advanced structure‐based and computational drug discovery technology. The company’s platform combines the latest advancements in visualization of molecular interactions, computational chemistry, and data integration to design orally available, superior small molecule medicines that overcome current limitations of biologic and peptide drugs. We are advancing a clinical‐stage pipeline of differentiated treatments focused on chronic diseases with high unmet need, including cardiovascular, metabolic, and pulmonary conditions. With offices in California and Shanghai, Structure Therapeutics has the benefit of being at the center of life science innovation in both the US and China and capitalizing on the strengths of each geographic location. Position Summary: Structure Therapeutics is seeking a Director of AI Engineering to build and scale production AI capabilities across drug discovery, clinical development, manufacturing, and business operations. This is a hands-on player-coach position. Approximately 60-70% of the role will involve direct technical contribution: writing and reviewing code, developing AI applications, building model and data pipelines, creating evaluation frameworks, troubleshooting deployments, and converting prototypes into reliable production systems. The remaining time will focus on technical direction, team development, and close partnership with scientific, product, data, and business leaders. The Director will lead a focused team while personally contributing to its most important systems. This role is intended for a leader who can move comfortably between AI strategy, system architecture, scientific problem-solving, and implementation. Job Responsibilities: Production AI Applications

• Personally design, code, test, and deploy production AI applications, services, APIs, and reusable engineering components.
• Build generative-AI solutions for scientific search, clinical-study design, document generation and classification, knowledge retrieval, workflow automation, and decision support.
• Develop retrieval-augmented generation, semantic search, tool-calling, structured-generation, and agentic workflows.
• Fine-tune or adapt language and machine-learning models when general-purpose models do not meet domain requirements.
• Build human-in-the-loop review and validation workflows for scientific, clinical, and regulated use cases.
• Develop document-intelligence capabilities that extract, classify, validate, generate, and route regulated content.
• Prototype directly with scientists and domain experts, then productionize the solutions that demonstrate meaningful value. AI Platform Engineering
• Build and evolve a centralized AI platform that enables teams across Structure to develop and deploy secure, reusable AI capabilities.
• Create shared services for model access, retrieval, prompt and workflow management, identity, authorization, observability, and evaluation.
• Design systems that support commercial foundation models, cloud AI services, specialized scientific models, and open-source models.
• Build scalable model-serving and data-processing capabilities for language, document, imaging, and structured-data applications.
• Develop production APIs and integration patterns that connect AI services with scientific, clinical, and enterprise systems.
• Optimize system performance, inference latency, reliability, and cost.
• Maintain clear abstraction layers so models and vendors can change without requiring complete application rewrites. LLMOps, MLOps, and Software Delivery
• Implement the complete AI lifecycle, from data ingestion and experimentation through validation, deployment, monitoring, and retirement.
• Establish CI/CD processes for AI applications, models, prompts, retrieval configurations, and infrastructure.
• Build automated evaluations covering accuracy, groundedness, relevance, safety, latency, cost, and domain-specific performance.
• Create regression tests and release criteria for AI systems used in scientific or regulated workflows.
• Implement monitoring for model behavior, data drift, retrieval quality, system performance, and user feedback.
• Maintain versioning and traceability across training data, models, prompts, code, and evaluation results.
• Turn experimental notebooks and proofs of concept into tested, documented, observable, and supportable production services. Clinical and Scientific AI
• Partner with Clinical Development and Clinical Operations teams to automate labor-intensive elements of study design and study build.
• Develop systems that improve the quality and efficiency of clinical metadata, edit checks, test-data generation, validation, and study documentation.
• Build AI-assisted workflows for informed-consent content and other patient-facing or regulated documents.
• Develop document-classification and intelligent-filing capabilities for clinical and regulatory content.
• Collaborate with research teams on scientific knowledge retrieval, computational modeling, multimodal data, imaging, and biomarker-oriented applications.
• Apply AI to high-value workflows while preserving appropriate scientific and human oversight. Responsible AI, Security, and Compliance
• Build access controls, audit trails, validation evidence, and human-review mechanisms directly into AI systems.
• Partner with Quality, Legal, Compliance, Security, IT, and Data teams to protect regulated data and proprietary scientific information.
• Support applicable GxP expectations, including controlled change, traceability, reproducibility, and auditability.
• Establish practical standards for model validation, approved use, monitoring, documentation, and incident escalation.
• Design controlled environments for evaluating models and vendors using representative, production-relevant data.
• Ensure external services receive only authorized information and that sensitive data is handled according to company policy. Hands-On Technical Leadership
• Lead and mentor AI and machine-learning engineers while remaining an active contributor to the production codebase.
• Participate directly in architecture design, coding, code review, testing, deployment, and incident resolution.
• Establish engineering standards for maintainability, documentation, reproducibility, security, and operational ownership.
• Break ambiguous scientific and business problems into testable technical milestones.
• Build the team thoughtfully, developing both AI depth and strong software-engineering practices.
• Create an environment in which engineers can experiment quickly while maintaining production and regulatory discipline.
• Communicate technical decisions clearly to scientists, engineers, product leaders, and executives. Technology and Vendor Evaluation
• Conduct hands-on evaluations of foundation models, cloud AI services, scientific software, development frameworks, and infrastructure providers.
• Build benchmarks based on realistic Structure workflows instead of relying primarily on vendor demonstrations.
• Compare alternatives based on output quality, validation requirements, security, integration effort, scalability, latency, cost, and operational risk.
• Make pragmatic build-versus-buy recommendations and retain sufficient internal capability to avoid unnecessary vendor dependence. Qualifications:
• BA/BS in Computer Engineering, MBA a plus.
• Eight or more years of combined experience in AI, machine learning, software engineering, data science, or computational research.
• Demonstrated experience building AI or machine-learning capabilities that progressed from experimentation into production use.
• Current hands-on proficiency in Python and SQL, including testing, debugging, profiling, API development, and code review.
• Experience building generative-AI or LLM applications using retrieval, structured gene