Distributed Training & Inference Optimization Engineer (LLM) - GPU Optimization Department (GPUOD)
Rakuten Crimson House, Japan
JPY 8000K-10800K (estimate) Mid-level Full Time
Tasks
- Collaborate on GPU cluster scheduling and resource allocation
- Develop benchmarking tools for training throughput and inference latency
- Implement inference optimizations for low latency and high throughput
- Optimize LLM training frameworks for GPU utilization and training time reduction
- Profile and optimize distributed training bottlenecks
- Research and apply LLM performance optimization techniques
- Tune LLM inference serving performance
Perks/Benefits
- N/A
Skills/Tech-stack
CUDA | Continuous batching | DeepSpeed | Dynamic batching | FSDP | KV caching | Kubernetes | Megatron-LM | Mixture of Experts | NCCL | Nsight | Profiling | PyTorch | Quantization | Speculative decoding | TensorBoard | TensorRT-LLM | Triton | VLLM
Education
Related jobs
-
AIOps | Artificial Intelligence | Big Data | Big data analytics | CUDASenior-level Full TimeOsaka, Japan1d ago
-
AIEducation | AIOps | Big Data | Big data analytics | CUDAOnsite work | Relocation support | Visa sponsorshipSenior-level Full TimeKawasaki, Miyagi, Japan1d ago
-
AIOps | Big Data | Big data analytics | CUDA | Computer VisionOnsite work | Relocation support | Visa sponsorshipSenior-level Full TimeNagoya, Aichi, Japan1d ago
-
AIOps | Artificial Intelligence | Big Data | Big data analytics | CUDAInternational projects | Onsite work | Relocation support | Visa sponsorshipSenior-level Full TimeYokohama, Kanagawa, Japan1d ago
-
AIOps | Big Data | Big data analytics | CUDA | Computer VisionOnsite work Japan | Relocation support | Visa sponsorshipSenior-level Full TimeSaitama, Japan1d ago
-
AI Pipelines | AWS | Airflow | Amazon EKS | Azure AKSJapanese language proficiency support | On site work Tokyo Japan | Relocation support | Visa sponsorshipMid-level Full TimeSaitama, Japan1d ago
-
AI Pipelines | AWS | Airflow | Amazon EKS | Azure AKSOn site Tokyo work | Visa sponsorshipMid-level Full TimeYokohama, Kanagawa, Japan1d ago
-
AI Pipelines | AWS | Airflow | Amazon EKS | Amazon SageMakerOn site Tokyo Japan | Relocation assistance | Visa sponsorshipMid-level Full TimeTokyo, Kantō, Japan1d ago
-
AKS | AWS | Airflow | CI/CD | Cloud platformBusiness level Japanese proficiency | Onsite work in Tokyo Japan | Relocation assistance | Visa sponsorshipMid-level Full TimeNagoya, Aichi, Japan1d ago
-
AI Pipelines | AWS | Amazon EKS | Amazon SageMaker | Apache AirflowOn site work Japan | Relocation support | Visa sponsorshipMid-level Full TimeOsaka, Japan1d ago
-
AI Engineer JPY 8000K-15000KAlloyDB | Anthropic | Cloud Monitoring | Cloud SQL | DatadogConference support | Flexible schedule | Learning SupportExecutive-level Full TimeTokyo R1d ago
-
Software Engineer - Data Infrastructure JPY 8600K-9700KAirflow | Apache Beam | Apache Flink | Apache Iceberg | Apache KafkaMid-level Full TimeJP - Tokyo, Japan1d ago
-
AWS | C++ | Cloud platform | Computer Vision | Data PipelinesOn-site workMid-level Full TimeTokyo, Kantō, Japan2d ago
-
3D Geometry | 3D Reconstruction | Activity Recognition | Camera Calibration | Computer VisionSenior-level Full TimeTokyo2d ago
-
A/B | A/B Testing | B testing | CI/CD | ContainerizationHybrid remote workSenior-level Full TimeShibuya, Japan R3d ago
-
Sales Engineer - Enterprise JPY 9815K-12000KAI | AWS | Application Performance Monitoring | Application performance | Cloud platformCareer pathing | Community guilds and inclusion talks | Employee stock purchase plan | Hybrid work | Mentor and buddy programMid-level Full TimeTokyo, Japan R3d ago
-
Backend Engineer, Agent Development - AI Lab (AILAB) JPY 8000K-14000KA/B | A/B Testing | Async architecture | B testing | CI/CDSenior-level Full TimeRakuten Crimson House, Japan3d ago
-
CT-NITRO-Experimental AI Engineer JPY 8600K-10000KAutoencoders | Deep learning | Docker | Fine Tuning | GitSenior-level Full TimeKochi, KL, IN, 6823133d ago
-
Forward-Deployed Data Scientist JPY 8000K-10250KAirflow | CI/CD | CatBoost | Code review | ETLDisability insurance | Donation matching | Employee stock purchase plans | Equal Paid Parental Leave | Flexible paid time offMid-level Full TimeTokyo6d ago
-
APIs | Agent systems | Cloud platform | Cost Optimization | CrewAIMid-level Full TimeTokyo, Japan6d ago
-
Senior AI Engineer - LLM Adoption Department (LLMAD) JPY 8600K-14000KAPI Integration | Agentic AI | Backend Development | Caching | Context engineeringSenior-level Full TimeRakuten Crimson House, Japan6d ago
-
DevOps Engineer - Conversational Experiences Department (CXD) JPY 6000K-10000KAWS | Ansible | Azure | Docker | ELKEntry-level Full TimeRakuten Crimson House, Japan9d ago
-
AI Engineer JPY 7500K-12000KChatGPT | Cloud PubSub | Cloud Storage | Datadog | Deep learningBook purchase support | Conference participation support | Full remote | Learning SupportMid-level Full TimeTokyo R11d ago
-
Software Engineer - Experimentation System - Mercari JPY 8600K-9700KA/B | A/B Testing | API Design | Allocation Optimization | B testingFlexible work hours | No core timeSenior-level Full TimeMinato City, Tokyo, Japan13d ago
-
Senior Data Scientist - AI Agent Development Section, Business Support Development Department (BSD) JPY 8000K-12500KAWS | Cloud Computing | Cloud platform | Data Analysis | Deep learningBusiness-level language support | Mentoring opportunitiesSenior-level Full TimeRakuten Crimson House, Japan14d ago