Network Bachelor

How to use this workbook Five projects build one stack, from an LLM that writes Dockerfiles to GPU inference that scales to zero. Do them in order: each project reuses something the previous one produced, and each ends with checkpoints you tick only after you have verified them on a real cluster. Every project follows

Part 6 of “vLLM in 2026: An Infrastructure Engineer’s Field Guide” For years, training and inference were separate worlds with separate clusters and separate teams. That boundary is dissolving. Reinforcement learning turns your inference engine into part of the training loop. Speculative decoding needs draft models that must be trained against the target you serve.

Explore LLM quantization with vLLM. Compare PTQ vs QAT, FP8, INT4, KV cache compression, and GPU memory optimization for efficient AI inference.

Explore agentic AI capacity planning with vLLM. Learn how prefix caching, KV offloading, and GPU optimization improve inference efficiency.

Explore distributed LLM inference, GPU networking, vLLM, expert parallelism, KV cache transfers, and Kubernetes-native AI serving with llm-d.

Compare vLLM multi-hardware inference across NVIDIA, AMD, Google TPU and Intel for performance, portability, cost and enterprise AI infrastructure. Today, vLLM multi-hardware inference can run across NVIDIA GPUs, AMD Instinct accelerators, Google TPUs, and Intel hardware

Learn how vLLM handles LLM inference using PagedAttention, KV cache, continuous batching, chunked prefill, torch.compile and production monitoring.

Build a sovereign LLM inference platform with Kubernetes, NVIDIA H200 GPUs, vLLM, NIM, LiteLLM and F5 for secure enterprise AI infrastructure.

Learn how retina scan security works, the privacy risks of biometric authentication, template protection, anti-spoofing controls, and biometric data protection.

Learn secure data destruction methods based on NIST SP 800-88 Rev. 2, including data sanitization, cryptographic erase, SSD sanitization, degaussing, and physical destruction.