Principal Software Engineer, Riot Games

• 10+ years of experience in software engineering, with substantial time spent in platform or infrastructure teams
• Proven technical leadership in building large scale distributed systems, production ML systems or model serving infrastructure at scale
• Deep experience with cloud-native systems (e.g., Kubernetes, containerization, autoscaling, observability stacks)
• Experience with one or more inference serving frameworks (e.g., NVIDIA Triton, KServe, TorchServe, BentoML, Seldon Core etc)
• Familiarity with GPU orchestration, performance tuning, and cost-aware scheduling
• Strong background in CI/CD automation, IaC tools (e.g., Terraform), and artifact management
• Hands-on experience with Python ML ecosystems, package management (Poetry, Conda etc), and vulnerability scanning
• Ability to mentor engineers, write clear documentation, and influence cross-functional stakeholders

Desirable

• Riot Engineers bring deep knowledge of specific technical areas but also value the chance to work in many broader domains
• As a Software Engineer, you’ll also dive into projects that focus on team cohesiveness and cross-team goals
• You’ll lead without authority and provide other engineers with a clear illustration of extraordinary engineering
• As the Principal Software Engineer on the ML Platform team, you’ll embrace MLOPs concepts– architecting, building, and leading core infrastructure for ML model deployment, observability, and lifecycle management
• This role is foundational in bringing modern ML inference practices to our game and platform teams—reducing time-to-deploy, reducing effort of operations, avoiding a cycle of reinventing the wheel, scaling infrastructure intelligently, and enabling data scientists, ML engineers, and product teams to move faster leveraging production grade ML with confidence
• You’ll design for both GPU and CPU workloads, live testing (A/B and shadow modes), model versioning—creating a platform that is robust, cost-efficient, and extensible for Riot’s present and future needs
• Your work will directly enable high impact ML use cases across Riot’s games and internal teams
• This engineer will be a part of founding a new ML Platform team and serve as its first technical leader— writing initial critical systems code, and guiding long-term strategy and direction as the team grows
• You'll report directly to the Director of Data Science
• Architect and implement Riot’s core ML inference infrastructure, with a focus on both live inference and nearline batch inference for scalable model serving, CPU and GPU-aware orchestration, and automated deployment pipelines
• Partner with researchers, game teams, and platform engineers to understand product needs and deliver generalizable, reusable solutions
• Define and build CI/CD workflows for ML artifacts—supporting rapid iteration and safe promotion from dev to production and MLOps practices
• Own tooling for environment and dependency management strategies (e.g., Conda/Poetry lock files, secure image builds) for ML runtimes
• Instrument and emit platform metrics for observability, model monitoring, drift detection, CPU/GPU utilization, and latency SLAs
• Establish patterns and tooling for multi-version model support, blue/green and shadow deployments, and rollback
• Be thoughtful on developer UX and incorporate an iterative approach to improving
• Serve as the technical founding voice for a new platform—defining long-term architecture, mentoring incoming engineers, and collaborating on hiring
• Contribute upstream to shared infra initiatives and build a feedback loops and collaboration models with other Riot platform teams
• For this role, you'll find success through craft expertise, a collaborative spirit, and decision-making that prioritizes the delight of players