model-serving 11
- LLM Inference Parallelism Is a Rank-Topology Problem
- Signal-Decision Architecture: Multi-Dimensional Semantic Routing for vLLM Clusters
- Architecting Production ML Inference: From Microservices to vLLM Clusters
- Kubernetes for Machine Learning: GPU Scheduling, NVIDIA NIM, and KEDA Autoscaling
- The Production LLM Stack: Architectural Blueprint Across Seven Engineering Layers
- LLMOps Architecture: End-to-End Streaming, QLoRA Tuning, and Serving
- Engineering Production LLMs: The 4-Pipeline Architecture of the LLM Twin System
- The Production LLM Technology Stack: Architecture Patterns from Ingestion to Observability
- NVIDIA NIM on Hugging Face Endpoints: Containerized Inference and TensorRT Acceleration
- Llama 3 70B on AWS Inferentia2: Optimum Neuron, NeuronCore-v2, and TGI Serving
- The Open LLM Curriculum: Deconstructing Scientist and Engineer Competency Maps