Reading paths

Technical essays arranged in sequence to study AI architecture, context, tools, and evaluation.

AI systems: architecture to evaluation

Choose an architecture, decide how to provide context, connect tools, and evaluate reliability.

  1. AI Agents vs. Workflows: When to Use Each Architecture

    A practical framework for choosing between deterministic AI workflows and autonomous agents based on task structure, reliability, cost, latency, and control.

  2. RAG vs. Fine-Tuning vs. Tool Use: A Practical Decision Framework

    A practical framework for choosing between retrieval-augmented generation, fine-tuning, and tool use based on knowledge freshness, behavior, actions, latency, and control.

  3. Model Context Protocol (MCP) Explained: What It Solves and What It Does Not

    A practical explanation of Model Context Protocol: what MCP standardizes, how tools/resources/prompts fit together, where it differs from function calling and APIs, and what it does not solve.

  4. How to Evaluate AI Agents: Reliability, Cost, Latency, and Failure Modes

    A practical framework for evaluating AI agents across task success, reliability, cost, latency, tool use, trajectories, and failure modes before production.