Introducing our team (project). 

• KRAFTON's AI Research division builds the GPU infrastructure required for model development, and develops and operates AI platforms based on Kubernetes and Slurm.
• Covering comprehensively GPU cluster construction and operation, model training and experimentation environments, ML pipelines, infrastructure automation and observability, we are building an environment in which the AI workloads needed for game production and services can run stably and efficiently.
• This position is a hands-on senior engineer role that operates and advances the large-scale GPU infrastructure already built, while designing the structure of infrastructure to be expanded in the future and developing it so that research, development and service organisations can use it stably and efficiently.
• Also, based on practical experience and technical insight, you will carry out technical review and make proposals on the future direction of computing infrastructure expansion, and take part in proposing and executing KRAFTON's GPU and computing operations strategy.

Introducing the mission to undertake with our team.

• This position is a hands-on senior engineer role that, while jointly operating and advancing large-scale GPU infrastructure, enables the research, development and service organisations to use it stably and efficiently.
• You will directly take part in operational stabilisation, performance improvement, resource efficiency and operational automation of large-scale GPU infrastructure based on Blackwell Ultra.
• You will design, build and operate scheduling, multi-tenancy, workload isolation, quotas, observability and incident response systems for the Kubernetes-based ML/GPU platform.
• Based on the characteristics of training and inference workloads, you will establish operational strategies that improve GPU utilisation, waiting time, throughput and cost efficiency and apply them to the actual platform.
• Based on GPU capacity planning and utilisation analysis, you will propose technical review of future computing infrastructure expansion direction and whether to use external infrastructure, along with technical judgement.
• You will advance a reproducible operational system based on GPU and coordinate the requirements of multiple teams from a common platform perspective.

We want to grow together with someone who has these experiences! (Required)

• Experience in designing, building and operating AI/ML infrastructure based on large-scale GPU clusters or Kubernetes and Slurm
• Experience in understanding overall computing infrastructure, including GPU servers, networking and storage, and building a stable operating environment
• Experience applying GPU resource management, scheduling, multi-tenancy, workload isolation and quotas to production environments
• Experience in analysing GPU utilisation, performance, stability and cost to optimise infrastructure
• Experience in analysing failures and performance issues from a system perspective and carrying out root cause resolution and structural improvement
• Experience in collaborating with research, development and service organisations to define infrastructure requirements and propose and execute expansion and operational direction
• No obstacles to overseas business travel

If you have these experiences, you are the person we are looking for! (Preferred)

• Experience in establishing cost and performance optimisation strategies for GPU infrastructure and applying them to operations
• Experience in collaborating with research, development and service organisations to define infrastructure requirements and propose technical alternatives and execution priorities
• Experience in reviewing and executing computing infrastructure expansion strategies including on-premises, cloud and external infrastructure
• Understanding of the overall ML workflow, including model training, experimentation, deployment, inference and serving
• Experience in building repeatable and efficient infrastructure operating systems using IaC, automation, CI/CD and observability tools
• Experience in improving operational efficiency and automation productivity using generative AI, LLM-based tools and code assistants

To take on KRAFTON's challenges with us, the following selection process is required.

• Application review > Preliminary interview (Phone Interview) > *Assignment > Personality assessment (Personality Assessment) > Technical fit interview (Technical Fit Interview) > Comprehensive interview (2-3 rounds) (Culture Fit Interview) > Offer and joining
• *This process may vary depending on each position and candidate.
• This is a rolling recruitment and may close early once the position is filled.
• Interviews and job tests may be added if necessary.
• Selection schedule and results (pass or fail) will be notified individually by the email or phone number you registered in your application.

Please check the required documents!

• Application form (free format), cover letter, career description, portfolio (if necessary)
• For entry-level candidates, please focus on the cover letter; for experienced candidates, please focus on the career description.
• When attaching a portfolio, please check the guidance below. 
  

Work location

• Yeoksam Centerfield

Employment type 

• Permanent (however, employment type may change depending on the result of salary and conditions negotiation with the candidate)