Senior Applied AI Engineer, Cybersecurity

The Cyber Defense Applied AI team is building NVIDIA’s agent-first security operations. We develop and operationalize trusted AI agents that augment analyst judgment, automate security work, and improve the efficiency of detection, investigation, and response processes. We combine NVIDIA AI technologies with open models, frontier models, and strategic partner capabilities to apply the best approach to each security problem.   As a Senior Applied AI Engineer, Cybersecurity, you will build AI systems that perform real security work. You will develop agents that reason over security telemetry and organizational context, use security tools, and support investigation and response. You will take capabilities from experimentation through evaluation, optimization, deployment, and production operation. You will also assess emerging approaches,   adapt   what already works, and build   new solutions   where meaningful gaps   remain . This role carries significant technical autonomy and influence. You will make evidence-based decisions about what to build, adopt, integrate, or develop with partners and use operational results to shape the Applied AI roadmap!

Partner with security practitioners to identify high-impact workflows and lead the delivery of agentic systems that improve analyst decision-making and accelerate detection, investigation, and response.     ​ Provide technical direction for complex agentic AI initiatives, shaping architecture, project goals, and engineering decisions across teams. Drive work from ambiguous problems to measurable operational outcomes.    Build and develop context-aware agents that analyze security data streams and institutional knowledge, use approved tools, and support greater autonomy as operational evidence and controls allow.     Establish repeatable evaluation for models and agents using realistic security environments, curated datasets, analyst ground truth, and task-specific benchmarks. Evaluate end-to-end behavior through automated scoring, trajectory analysis, and adversarial testing.  Use evaluation results, production traces, and analyst feedback to improve agent quality, reliability, and efficiency.   Optimize   models, retrieval, context, orchestration, and inference against measurable security outcomes.    Take AI capabilities from experimentation to production using strong software engineering and   MLOps / LLMOps   practices. Build continuous evaluation, observability, versioning, controlled deployment, and safe rollback into the lifecycle.   Evaluate NVIDIA AI technologies alongside open-source, frontier, and strategic partner capabilities within an interoperable, multi-model approach. Make evidence-based recommendations on what to adopt, adapt, build, integrate, or co-develop.    Translate technical findings into clear recommendations that influence architecture, Applied AI priorities, and partner roadmaps. Turn proven approaches into reusable capabilities that strengthen NVIDIA and the broader open, interoperable AI security ecosystem.   What we need to see: BS, MS, or PhD in Computer Science, Artificial Intelligence, Machine Learning, Software Engineering, Cybersecurity, or a related technical field, or equivalent experience.   8+ years of relevant experience building and shipping AI, machine learning, or intelligent software systems, including technical ownership of complex production initiatives.   Strong software engineering skills, particularly in Python, with experience building production systems using languages such as TypeScript or C#. Demonstrated ability to design reliable and scalable systems beyond prototypes or experimental notebooks.   Hands-on experience designing and developing modern AI systems using large language models, retrieval-augmented generation, agentic architectures, agent harnesses, or related approaches.   Experience designing AI evaluations and benchmarks using curated datasets, ground truth,