AI

Distinguish model knowledge, short-term context, and long-term memory, with an emphasis on storage, retrieval, and lifecycle management.

Organize practical AI evaluation around criteria, measurement methods, model selection, and evidence-based decisions.

Survey the dataset engineering lifecycle: requirements, acquisition, synthesis, cleaning, validation, and quality control.

Connect gateways, context engineering, guardrails, caching, agents, and observability with a user-feedback loop.

Study inference optimization through latency, throughput, memory use, and cost across models, runtimes, and serving systems.

Map the AI technology stack into four layers: compute infrastructure, models, application platforms, and AI applications.