Software Engineer - ML Infrastructure

Gray Swan AI•Pittsburgh, PA

About The Position

Gray Swan is hiring a Software Engineer, AI Infrastructure to work across machine learning, software engineering, and production infrastructure. The engineer will partner closely with the ML team to build systems, tools, and workflows that move models from experimentation into reliable production environments. The role focuses on improving the developer experience around model training and evaluation, building scalable infrastructure for ML workflows, and translating customer requirements and policy considerations into practical technical solutions. The position suits a software engineer who enjoys complex infrastructure challenges and close collaboration with ML researchers and engineers to accelerate development and deliver real-world impact. Gray Swan works to empower safe and secure AI use, evaluating AI models for frontier labs and building real-time threat detection and adaptive adversarial red-teaming agents. The company has approximately 70 people and is well-funded and growing quickly.

Requirements

Experience building and maintaining production-grade systems, services, or infrastructure.
Experience with Python and comfort working across backend systems, APIs, and infrastructure code.
Familiarity with cloud infrastructure, distributed systems, containerization, or platform engineering.
Knowledge of ML workflows, including model training, inference, evaluation, or deployment.
Ability to collaborate with ML engineers and researchers to translate complex technical requirements into reliable software.
Ability to take ownership, navigate ambiguity, identify problems independently, and deliver practical solutions in a fast-moving environment.

Nice To Haves

Experience building ML infrastructure, ML platforms, or internal developer tooling.
Familiarity with Kubernetes, Docker, and cloud platforms such as AWS or GCP.
Experience with ML infrastructure tools such as Ray, Kubeflow, MLflow, or Weights & Biases.
Experience with model serving, distributed training, inference infrastructure, or large-scale evaluation pipelines.
Exposure to AI safety, AI security, adversarial testing, or model behavior evaluation.
Experience translating customer requirements, security constraints, or policy considerations into technical implementations.

Responsibilities

Build and maintain infrastructure, services, and deployment workflows that move ML models from experimentation into reliable production environments.
Develop internal tools and automation to improve the developer experience for model training, experimentation, evaluation, and iteration.
Design and optimize systems for model evaluation pipelines, experiment orchestration, reproducibility, and performance monitoring.
Partner with ML engineers and researchers to identify infrastructure bottlenecks and build solutions that accelerate model development and deployment.
Support customer-specific technical requirements and policy-driven workflows by translating evolving needs into practical, scalable engineering solutions.
Improve the reliability, scalability, and observability of systems powering ML capabilities.