LLM Engineer (LLM Training)

42dot
South Korea
Workplace: HybridFull timeFunction: Education & TrainingExperience: 3+ yearsSkills: ["Problem-solving","Research and development","Troubleshooting","Quality improvement"]

Design and improve LLM training pipelines that produce high-quality, reliable generative language models for real services. Drive efficiency improvements across pre-training and post-training, and enhance accuracy and stability of generation through learning structures that support Self-Refine. Work on model design, training, evaluation, optimization, and GPU-based troubleshooting using distributed training frameworks, contributing to continuous quality gains.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
42dot
42dot
1 month ago

LLM Engineer (LLM Training)

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 57 minutes agoStatus: Live

Job Summary

Design and improve LLM training pipelines that produce high-quality, reliable generative language models for real services. Drive efficiency improvements across pre-training and post-training, and enhance accuracy and stability of generation through learning structures that support Self-Refine. Work on model design, training, evaluation, optimization, and GPU-based troubleshooting using distributed training frameworks, contributing to continuous quality gains.
Location: South Korea
Workplace: Hybrid
Employment Type: Full time
Job Function: Education & Training
Seniority: Mid level

Key Responsibilities

  • •Improve the efficiency of the LLM training process, including pre-training and post-training.
  • •Enhance overall accuracy and stability of generated outputs.
  • •Design a learning structure that enables Self-Refine for better generation quality.

Key Requirements

  • •3+ years of experience in Deep Learning or NLP (open to master’s new grads).
  • •Broad R&D experience related to Deep Learning or NLP.
  • •Strong programming skills (Python & PyTorch).
  • •Experience with model design, training, evaluation, and optimization using PyTorch.
  • •Experience training LLMs with GPUs and troubleshooting, including distributed training frameworks (e.g., Slurm, DDP, Horovod).
Experience:3+ yearsDeep learningNLP
Skills:Problem-solvingResearch and developmentTroubleshootingQuality improvement
Tech Stack:PythonPyTorchGPUSlurmDDPHorovodDockerKubernetesLLMSelf-RefineSupervised Fine-TuningParameter Efficient Fine-TuningTrainingInference

Company Brief

42dot
Develops autonomous driving and mobility software platforms, including AI-based perception, mapping, routing, and connected-vehicle technologies. It works on next-generation transportation systems and self-driving vehicle capabilities for automotive applications.
Industry: Autonomous Vehicles
Company Size: Large (251 to 1,000 employees)
Growth: Scaleup
Headquarters: Seoul, South Korea
Founded: 2019
WebsiteLinkedIn