AI engineering is the work of building, serving and operating applications on top of foundation models. It sits between machine learning research and software engineering: prompts and context, retrieval, agents and tools, inference and cost, evaluation and testing, observability and guardrails. This hub groups the related pages on this site into seven areas. Each card links to a guide or directory that is updated on its own page.
Building Blocks: Prompts, Context, RAG and Models
14 entriesThe core techniques and components of an AI application: prompts and context, retrieval, embeddings and rerankers, fine-tuning, and the stores behind them.
Agents and AI-Assisted Coding
9 entriesFrameworks, protocols and tools for building agents, and the coding agents engineers use day to day.
Inference, Serving and Cost
11 entriesRunning models in production: inference engines and platforms, gateways and routers, and optimization of latency, memory and tokens.
Evaluation and Testing
5 entriesHow to know whether an AI system works: LLM and agent evaluation methods, test tooling and metrics.
Operations and Observability
7 entriesKeeping AI systems running: LLMOps and MLOps, monitoring, tracing and observability.
Safety and Guardrails
2 entriesControls that keep agents and LLM applications inside safe limits.
Learning Resources
4 entriesCurated repositories and roadmaps for learning the craft.
Related Pages
5 entriesRelated hubs and directories on this site.