12 Core Terms for AI Agent Engineering in 2026
PawelHuryn · x · 2026-08-09
The author outlines the 12 most critical terms in current AI Agent engineering and workflows to help PMs and developers align their understanding:
- Agent: Decides how the work gets done.
- Reasoning effort: How long the model thinks (the author noted that many models silently ignore this parameter when set via OpenRouter).
- Skill: Procedures your agent loads when needed.
- MCP: How agents plug into external products.
- Context engineering: Curating what the agent sees.
- AI evals: Scoring the agent based on real failures.
- Vibe coding: Prompting until the demo works.
- Agentic engineering: Agents building what you ship.
- Harness: The architecture running the model, tools, and loop.
- Loop engineering: Keeping the agent trying until the task is done.
- Graph engineering: Fixed steps and fixed branches.
- Intent engineering: Leading with strategic context rather than just control.
More from coding & agent
- Developer Recreates Classic iBeer App as an MCP Experiment — Puzzleheaded_Mine392 · 2026-08-09
- Lilian Weng's Deep Dive: The Key to Agent Self-Improvement Lies in Harness Engineering — aigclink · 2026-08-09
- Adversarial Testing Catches Silent Prompt Injection Regression in Doc Assistant — OpeningBird6240 · 2026-08-09
- Why Is There No 'App Store' for Independent AI Agents Yet? — mgsz_ · 2026-08-09
- Vibe Coding Accelerates Shipping, But Outpaces User Validation — Ubunta · 2026-08-09
- Building Production AI Agents: Linear Team Shares Core Lessons — petergyang · 2026-08-09