K

KRAFTON

Develops and publishes cross-platform games

AI Inference Platform & Operations Engineer

Applied Machine LearningConfirmed live in the last 24 hours
Full-timeSeoul, South KoreaNo salary listed

About the Job

Requirements

  • At least 5 years of engineering experience, including at least 3 years in infrastructure, site reliability engineering, platform engineering, or machine learning systems, with direct operation of production services where incidents affect users.
  • Experience designing and operating deployment, load balancing, scheduling, and autoscaling in Kubernetes or an equivalent orchestration environment.
  • Experience building observability using metrics, logs, and traces, and structurally improving root causes through on-call incident response and postmortems.
  • Experience implementing operational automation tools or platform features in code and managing infrastructure and deployments reproducibly with infrastructure as code, GitOps, and continuous integration and delivery.
  • Experience aligning requirements with researchers, product developers, and other disciplines, and documenting and sharing design and operational decisions.
  • No restriction preventing overseas business travel.

Responsibilities

  • Build and operate the serving platform and control plane on Kubernetes-based GPU clusters, connect serving engines such as vLLM and SGLang with orchestration layers, and implement request routing, autoscaling, and multi-model operations.
  • Build and improve model performance and quality regression tests, deployment automation, canary and rollback pipelines, and API gateway authentication, quotas, and rate limits; create tools and guides for consistent model onboarding by business squads and model developers.
  • Monitor latency, throughput, availability, and error rates against service-level objectives agreed with squads; analyze bottlenecks and incident causes, participate in incident response and postmortems, and implement resulting operational improvements.
  • Operate GPU allocation, isolation, and priority policies, and measure GPU utilization and per-request and per-token costs by model and workload.
  • Improve routing, scheduling, caching policies, and model integration code to meet service requirements; validate serving-engine features against real traffic and extend the engines where needed.
  • Validate optimization implementations provided by AI research teams or open source, such as quantized checkpoints and speculative decoding, against real workloads; check quality and performance regressions before production deployment.

Desired Qualifications

  • Experience operating large-traffic or low-latency real-time services, including games, and improving service stability in response to traffic changes.
  • Experience performing capacity planning for changing demand and operating quotas, priorities, and resource-isolation policies.
  • Experience serving LLM or machine-learning models in production on GPUs using vLLM, SGLang, TensorRT-LLM, or Triton Inference Server.
  • Understanding serving-engine scheduling, batching, and KV cache behavior, and applying continuous batching, prefill-decode disaggregation, prefix caching, or KV cache offloading to improve throughput or latency under real traffic.
  • Experience analyzing and improving performance bottlenecks in multi-node GPU environments or at the systems level.
  • Experience validating optimization results from research teams or open source, such as quantized checkpoints and speculative decoding, with quality and performance regression tests and deploying them to production.

About the Company

KRAFTON

Develops and publishes cross-platform games

Krafton is a group of independent game studios that develops and publishes video games for PC, console, and mobile. Its business model centers on delivering high-quality games and monetizing them through game sales, in-game purchases, and licensing deals, with PUBG as a flagship title. Krafton differentiates itself by coordinating a network of studios and investing in technology and publishing support rather than relying on a single in-house studio, enabling a diverse catalog and esports initiatives. Its goal is to push the boundaries of gaming and connect people through shared entertainment moments across platforms.

Company size
51-200
Company stage
IPO
Headquarters
Seongnam-si, South Korea
Founded
2007

Recently Posted Jobs

Explore more opportunities at KRAFTON

See more jobs