Strategies to reduce GenAI and agent token costs while maintaining task quality through practical controls and telemetry.
Deep dive into designing custom AI agent harnesses, covering architecture layers like loops, tools, context, and control for production systems.
Explores the concept of typed, budgeted, and auditable memory for AI agents, moving beyond flat retrieval to structured context management.
An overview of coding agent components, including tools, memory, and repo context, and how they enhance LLM performance in practice.
Explains the difference between an AI agent's inner loop (verifying work within a task) and outer loop (learning across tasks).
Explains the core concepts of AI coding agents (rules, commands, skills, etc.) and provides a unified mental model for understanding them.
Proposes a new AI agent architecture based on Alfred North Whitehead's process philosophy, treating agents as dynamic processes rather than static entities.
