AI Agents
Building with LLMs beyond single-turn prompts: multi-agent architectures, orchestration patterns, tool use via MCP, retrieval-augmented generation, and the failure modes that only appear in production agentic systems.
MCP, OpenAPI, and the Missing Runtime Layer
OpenAPI describes APIs. MCP lets LLM hosts discover and invoke tool providers at runtime. A practical systems explanation without the USB-C metaphor.
Multi-Agent AI System: Architectural Concept
Not a production blueprint, but a practical mental framework: how to decompose complex development into specialized roles, verification loops, and human control gates — with explicit limitations, failure modes, and real trade-offs.
RAG Architecture Patterns for Production: Chunking, Hybrid Search, and Access Control
How to build RAG that works in production: chunking strategies, hybrid search, RBAC/ABAC access control, evaluation, and latency optimisation. Visual guide.