Principal AI Researcher: Autonomous Network Modeling

Cisco; San Francisco (CA); On-site; USD 296,000 - 375,000; Full time

Job summary

Cisco's AI Software & Platform Group leads research to build world models that understand and predict network behavior to operate networks autonomously. This role focuses on world models, reinforcement learning, and model-based approaches for scalable, safe networking solutions.

You will work with engineers to evaluate real data and publish at top AI conferences. The team seeks experts who can take research from idea to prototype, collaborate across disciplines, and contribute to open-source

Qualifications

• Masters in Computer Science, Machine Learning, AI, Robotics, or related field with 8+ years of research experience.
• Research in world models, RL, model-based RL, representation learning, planning, or sequential decision-making.
• 3+ years building, training, and optimizing DL models using PyTorch/JAX or comparable frameworks; production/research-scale models.
• Track record of publications in leading AI/ML conferences or equivalent.
• Ability to take a research problem from idea to working prototype.

Responsibilities

• Develop world models that learn the state and dynamics of large-scale network environments.
• Build models predicting network behavior and impact of configurations.
• Develop model-based RL for network optimization, troubleshooting, remediation.
• Create simulation environments for training/evaluating models and agents.
• Explore representation learning, planning, and sequential decision-making for network environments.
• Evaluate reliability and safety of learned policies before deployment.
• Collaborate with networking engineers to evaluate on real data and publish research.

Skills

World models

Reinforcement learning

Model-based RL

Representation learning

Planning

Sequential decision

PyTorch

JAX

Research publications

Education

Master's degree in CS/ML

Tools

PyTorch

JAX

Job description

Cisco's AI Software & Platform Group leads research to build world models that understand and predict network behavior to operate networks autonomously. This role focuses on world models, reinforcement learning, and model-based approaches for scalable, safe networking solutions.

You will work with engineers to evaluate real data and publish at top AI conferences. The team seeks experts who can take research from idea to prototype, collaborate across disciplines, and contribute to open-source