| Company Name | K.P.P. Packaging Pte Ltd |
|---|---|
| Company UEN | 198600913R |
Model Deployment & Infrastructure • Deploy and configure LLMs on local infrastructure (on-premise servers, edge devices, private clouds) • Optimize model inference performance for resource-constrained environments • Implement containerization solutions (Docker, Kubernetes) for model serving • Design and maintain scalable ML inference pipelines • Monitor model performance, latency, and resource utilization Model Training & Fine-tuning • Fine-tune pre-trained LLMs for domain-specific use cases • Implement custom training pipelines for local model adaptation • Manage distributed training across multiple GPUs/nodes • Develop data preprocessing and augmentation strategies • Implement reinforcement learning from human feedback (RLHF) techniques Model Optimization & Compression • Apply model quantization, pruning, and distillation techniques • Implement knowledge distillation for model size reduction • Optimize models for specific hardware architectures (CPU, GPU, mobile) • Develop efficient attention mechanisms and caching strategies Integration & Development • Build APIs and microservices for model serving (REST, gRPC) • Integrate LLMs with existing applications and workflows • Develop prompt engineering strategies and templates • Implement retrieval-augmented generation (RAG) systems • Create monitoring and logging systems for production deployments Security & Compliance • Implement data privacy and security measures for local deployments • Ensure compliance with data protection regulations • Develop secure authentication and authorization mechanisms • Implement audit trails and model governance frameworks
Technical Skills • Programming Languages: Python (advanced), C++, CUDA • ML Frameworks: PyTorch, TensorFlow, Hugging Face Transformers • Model Serving: TensorRT, ONNX, OpenVINO, Triton Inference Server • Cloud/Container: Docker, Kubernetes, cloud platforms (AWS, GCP, Azure) • Infrastructure: Linux administration, GPU programming, distributed systems • APIs & Integration: REST APIs, gRPC, microservices architecture LLM-Specific Knowledge • Experience with LLM architectures (Transformer, BERT, GPT variants) • Knowledge of attention mechanisms, positional encoding, and tokenization • Familiarity with fine-tuning techniques (LoRA, QLoRA, adapters) • Understanding of inference optimization (KV cache, batching, pipeline parallelism) • Experience with model compression techniques (quantization, pruning, distillation) Education & Experience • Master's degree in Computer Science, AI/ML, or related field (PhD preferred) • 3+ years of experience in ML/AI development • 2+ years of specific experience with LLMs • Proven track record of deploying ML models in production environments Preferred Qualifications • Experience with open-source LLM frameworks (vLLM, Text Generation Inference) • Knowledge of federated learning and privacy-preserving ML • Experience with vector databases (Pinecone, Weaviate, Chroma) • Background in natural language processing and computational linguistics • Publications in top-tier ML conferences (NeurIPS, ICML, ICLR) • Experience with MLOps tools (MLflow, Kubeflow, Weights & Biases) Additional Skills • Strong problem-solving abilities and analytical thinking • Excellent communication skills for technical documentation • Experience with agile development methodologies • Understanding of software engineering best practices • Knowledge of version control (Git) and CI/CD pipelines
| Job Title | A/MLI Developer / Machine Learning Engineer |
|---|---|
| Salary | SGD3,000.00 - 4,000.00 |
| Employment Type | Full Time |
| Working Experience | 1 Years |
| Qualification | Post Graduate Diploma / Certificate |