below marketmost similar roles pay hereabove market
This role pays more than
60%
of similar roles. Most pay
$148,625–$238,384
— the blue band above.
At the midpoint, this role pays about
$211k
versus about
$194k
for comparable roles.
Based on 240 similar postings.
Employer
About Cisco
Cisco Systems is the world''s leading networking technology company, designing and manufacturing networking hardware, telecommunications equipment, and cybersecurity solutions for businesses and governments. Industry: Networking Technology & Cybersecurity
Cisco currently has
303 open roles
on FindRole.
Listed pay typically runs
$167,700–$245,200
across 218 roles with salary data.
The AI Researcher (Infrastructure) III joins the Infrastructure & Systems team to build the foundational platforms required for frontier AI research. This role involves designing, building, and operating systems that enable researchers to move from initial ideas to reproducible experiments and deployable models. Key responsibilities include developing shared platforms for pre-training, fine-tuning, reinforcement learning, and inference, while managing GPU and CPU computing infrastructure across clouds. You will create reproducible environments through automated container builds, manage workload orchestration, and improve observability across the stack to diagnose bottlenecks in GPU utilization and data loading. Required skills include Python, PyTorch, and production-grade software engineering. You must understand large language models, data pipelines, and scalable AI system design, with experience in distributed training frameworks like Megatron or NeMo.
Build shared platforms and developer tools for pre-training, fine-tuning, reinforcement learning, evaluation, and inference.
Develop and operate GPU and CPU computing infrastructure across various clouds and regions.
Create reproducible training and serving environments using automated container builds and dependency management.
Build reliable rollout collection and isolated execution environments for agentic workloads.
Connect experiment configurations, code, datasets, and checkpoints to ensure reproducible research runs.
Improve observability and performance by diagnosing bottlenecks in GPU utilization and distributed communication.
Partner with research teams to build dependable data and evaluation workflows.
Debug failures and deliver well-documented capabilities directly to research teams.
What we're looking for
Bachelor's + 5 years, Master's + 3 years, or PhD + 0 years of experience in Computer Science, AI, ML, or a related field.
Strong experience with Python, PyTorch, production-grade software engineering practices, and common AI/ML libraries.
Experience with large language models, model training, fine-tuning, evaluation, inference, and optimization.
Strong understanding of data pipelines, model experimentation, and scalable AI system design.
Strong communication skills to explain complex research and engineering concepts to cross-functional partners.
Experience with distributed GPU training or post-training systems, including parallelism strategies, checkpointing, recovery, and performance profiling (preferred).
Familiarity with frameworks such as Megatron, NeMo, VERL, or comparable distributed training and reinforcement learning systems (preferred).
Experience deploying and optimizing LLM serving systems such as vLLM or SGLang (preferred).