Principal AI Researcher, AISWP (Hybrid)

The application window is expected to close on: 10/04/2026 This is a hybrid position based out of Cisco's Seattle or San Jose office. Meet the Team The Cisco AI Research team is a dynamic group of scientists and engineers dedicated to building the future of proprietary intelligence. Our core mission is to develop purpose-built large language models tailored specifically for Cisco’s networking, security, and observability domains. We operate at the intersection of basic and applied research, ensuring that Cisco maintains its technological edge without reliance on external AI providers. Our team is composed of doers and experts who thrive in a collaborative environment, blending deep technical rigor with a shared passion for solving complex, real-world problems at a global scale.

As a Principal AI Researcher, you will define the technical vision and overarching architecture for Cisco’s proprietary intelligence, serving as a primary catalyst for how we secure and automate the global network. Moving beyond leading individual projects, you will define the multi-year research strategy, pioneering novel approaches in foundation model development, agentic workflows, and self-improving systems. You will bridge the gap between theoretical generative AI breakthroughs and hyperscale production deployments, guiding the evolution from generalized models to highly specialized, domain-expert architectures. By orchestrating complex, end-to-end pre-training and alignment initiatives, you will cement our technology at the absolute vanguard of the AI era, directly impacting the performance, security, and intelligence of the world’s digital backbone. Define and drive the overarching strategic research roadmap for purpose-built large language models securing and optimizing Cisco’s networking, security, data center, and observability infrastructure. Spearhead massive-scale pre-training, continued pre-training (CPT), and advanced reinforcement learning initiatives to establish industry-leading foundational models. Architect advanced agentic frameworks and autonomous operations, pushing the boundaries of reasoning capabilities and intelligent routing layers for IT environments. Serve as a principal technical advisor to executive leadership, translating abstract AI research breakthroughs into scalable, enterprise-grade product ecosystems. Elevate Cisco’s position as a premier global leader by publishing high-impact research, spearheading workshop proposals at top-tier AI venues, and driving the broader scientific discourse.

Bachelor’s degree in STEM with 15+ years of experience, a Master’s degree with 12+ years of experience, or a PhD in STEM or a related technical field with 8+ years of highly relevant research experience. Experience architecting and leading multiple full-cycle pre-training, continual pre-training, or reinforcement learning pipelines for massive-scale foundational models, dictating dataset curation, distributed training strategies, and alignment execution. Minimum of 6 publications in top-tier AI venues (ACL, EMNLP, ICLR, ICML, NAACL, NeurIPS), with a history of significant citations or influential contributions. 5+ years of experience engineering and scaling training pipelines using PyTorch or similar frameworks on massive distributed computing environments (e.g., cluster environments using 1000+ GPUs). 4+ years of experience pioneering the application of generative ML models to highly complex, domain-specific systems such as networking, cybersecurity, or data center infrastructure.

Demonstrated experience defining the technical architecture for AI Developer Experiences, agentic frameworks, or autonomous troubleshooting agents deployed in production environments. Deep expertise in advanced optimization methodologies, including context engineering, efficient serving architectures, and low-latency inference pipelines. Exceptional