AI Infrastructure Researcher - Scalable ML Platforms

Cisco Systems, Inc; San Francisco (CA); On-site; USD 184,000 - 239,000; Full time

Job summary

Foundation AI, within Cisco, is hiring an AI Researcher to design, build, and operate platforms that move ideas to reproducible experiments and deployable models. You will simplify launching complex AI workloads and scale experiments across teams.

You will own projects from identifying problems to delivering solutions, improving training stacks, scheduling, evaluation services, and making experiments easier to track and reproduce.

Qualifications

• Bachelor’s or higher in CS/AI/ML or related field with strong academic record.
• Proven experience with Python and PyTorch in production-grade software.
• Experience with LLMs, model training, fine-tuning, evaluation, and optimization.
• Understanding of data pipelines and scalable AI system design.
• Excellent communication for cross-functional collaboration.

Responsibilities

• Build shared platforms, services and tools for pre-training, fine-tuning, RL, evaluation and inference.
• Operate GPU/CPU infra across clouds and regions, with scheduling and resource sharing.
• Create reproducible training and serving environments via container builds and dependency management.
• Develop rollout and isolation environments for agent workloads with tracking and recovery.
• Link experiment configs, code, datasets, checkpoints, and results for reproducibility.
• Improve observability and performance across the stack, diagnosing bottlenecks.
• Partner with research and data teams on dependable data and eval workflows.
• Collaborate with researchers to understand workflows and deliver well-documented capabilities.

Skills

Python

PyTorch

Production‑grade software

AI/ML libraries

Data pipelines

Communication skills

Education

Bachelor's degree in CS/AI/related field

Master's degree in CS/AI/related field

PhD in related field

Tools

Megatron

NeMo

VERL

vLLM

SGLang

Docker

Job description

Foundation AI, within Cisco, is hiring an AI Researcher to design, build, and operate platforms that move ideas to reproducible experiments and deployable models. You will simplify launching complex AI workloads and scale experiments across teams.

You will own projects from identifying problems to delivering solutions, improving training stacks, scheduling, evaluation services, and making experiments easier to track and reproduce.