LLM Engineer (Optimization)
Tasks
- Analyze GPU memory and kernel performance
- Apply speculative decoding techniques
- Benchmark inference performance across backends
- Design serving architecture for cost latency throughput tradeoffs
- Develop GPU based LLM inference engine
- Develop model quantization pruning and distillation
- Improve memory efficiency for LLM serving
- Optimize LLM for edge and on device execution
- Optimize LLM inference latency
- Optimize LLM inference throughput
- Optimize LLM runtime for GPU and accelerators
- Profile GPU and runtime bottlenecks
- Use compiler and kernel optimization for inference
Perks/Benefits
- N/A
Skills/Tech-stack
Benchmarking | C++ | CUDA | Cache Compression | Inference Optimization | KV Cache Compression | KV cache | Knowledge Distillation | Kubernetes | Language Models | Large Language Models | MLIR | Model Pruning | Model Quantization | ONNX Runtime | Profiling | PyTorch | Python | Speculative decoding | TVM | TensorRT | Triton
Education
N/A
Related jobs
-
Applied AI Engineer - Finance Super App (Korea) KRW 44000K-44000KAPIs | Agent systems | Data integration | Evaluation | LLM APIsFast interview process | Remote workMid-level Full TimeSeoul, Korea22h ago
-
AI Deployment Engineer KRW 65000K-90000KAPI Integration | ChatGPT | Cloud Architecture | Enterprise Architecture | Generative AIHybrid work model | Relocation assistanceSenior-level Full TimeSeoul, South Korea1d ago
-
Bash | Cloud platform | Data Processing | Docker | Google CloudAsynchronous culture | Friendly work environment | Hands off management approach | Life-changing product impactMid-level Full TimeBusan, South Korea2d ago
-
Airflow | Apache Spark | Data Curation | Data Filtering | Data evaluationSenior-level Full TimePangyo (Software Dream Center), South Korea3d ago
-
Senior-level Full TimeSeoul5d ago
-
Software Engineer - Ads Experience (소프트웨어 엔지니어) KRW 65000K-90000KAPIs | Beam | BigQuery | Bigtable | C++Senior-level Full TimeSeoul, Korea5d ago
-
Deep Learning Engineer for Automated Driving KRW 48000K-67000KBehavior Prediction | C++ | Decision Making | Deep learning | Language ModelsMid-level Full TimeSeoul, Korea, Republic of6d ago
-
Accelerated computing | CAD | CUDA | Co-simulation | Contact and frictionSenior-level Full TimeKorea, Seoul6d ago
-
Cost Optimization | Data Preprocessing | Go | Hugging Face | Information RetrievalResearch publication opportunitySenior-level Full TimeGangnam District, Seoul7d ago
-
Bash | Cloud platform | Data Ingestion | Data Processing | DockerAsynchronous culture | Career impact on large scale product | Friendly work environment | Remote-friendly setupMid-level Full TimeSeoul, South Korea7d ago
-
Apache Airflow | Apache Spark | Data Lakehouse | Data Warehouse | Databricks WorkflowsMid-level Full TimePangyo (Software Dream Center), South Korea10d ago
-
Applied Scientist / Research Engineer KRW 44000K-50000KAgent systems | Data Curation | Distributed Training | GPU Computing | JAXFood allowance | Gym membership allowance | Health insurance | Transportation reimbursementEntry-level Full TimeSeoul10d ago
-
Agents | CI/CD | Data Curation | Data Pipelines | Deep learningSenior-level Full TimeSeoul10d ago
-
AWS | Apache Airflow | Apache Kafka | Apache Spark | Cloud platformSenior-level Full TimeSeoul10d ago
-
Applied AI, Machine Learning Engineer KRW 50000K-50000KAPIs | Back-end | Back-end development | Deep learning | Fine TuningEntry-level Full TimeSeoul10d ago
-
Applied AI Engineer, Fullstack KRW 44000K-44000KAlgorithms | Data Structures | Design Patterns | GenAI | JavaScriptMid-level Full TimeSeoul10d ago
-
Senior-level Full TimeSeoul10d ago
-
Apache Airflow | Azure Blob | Azure Blob Storage | BigQuery | Blob StorageBook purchase support | Conference attendance support | Device upgrade every 3 years | Employee accident insurance | Free snacksSenior-level Full TimeSeoul, Seoul, South Korea10d ago
-
D&A (Digitalization & Automation) Software Development Engineer (AI) KRW 65000K-90000KCause analysis | Data Governance | Data Pipelines | Data Quality | DatabricksSenior-level Full TimeHwasung Campus Building A, Korea, Korea, …10d ago
-
ADAS | AUTOSAR | Agent Frameworks | Autogen | Automotive validationEntry-level Full TimeKorea, Seoul10d ago
-
Checkpointing | Communication optimization | Data loading | Data parallelism | Deep learningEntry-level Full TimeSeoul11d ago
-
Intern- Deep Learning Researcher KRW 48000K-67000KAdversarial examples | Anomaly Detection | Deep generative models | Deep learning | Domain AdaptationEntry-level InternshipSeoul, Korea, Republic of11d ago
-
Intern- Deep Learning R&D for Automated Driving KRW 48000K-67000K3D Object Detection | C++ | Camera perception | Deep learning | LLMEntry-level InternshipSeoul, Korea, Republic of11d ago
-
Staff Database Engineer (Customer Success) KRW 65000K-90000KAWS CloudWatch | AWS IAM | AWS VPC | Amazon Aurora | Amazon RDSSenior-level Full TimeSeoul, South Korea12d ago
-
Associate/Experienced Embedded Software Engineer KRW 65000K-90000KARINC 429 | ARINC 664 | ARM | Architecture Fundamentals | BashSenior-level Full TimeKOR - Seoul, South Korea, Korea, …12d ago