---
title: Agentic Engineering
description: "You already ship code with Claude Code, Codex or Cursor. This course gives you the discipline behind it: context engineering, harness engineering, loop engineering, orchestration and evals, in clear terms and with runnable examples. Every code sample is a small harness written in plain Python and was executed with the real interpreter when this page was built, so every output you see is real."
author: Asem Abdo
course: Agentic Engineering
lesson: Start here
url: https://openplanr.dev/learn/agentic-engineering
updated: 2026-10-17
related:
  - https://openplanr.dev/learn/agentic-engineering/gently.md
---

# Agentic Engineering

By [Asem Abdo](https://www.linkedin.com/in/asemabdo/) · Updated 2026-10-17

*A free, interactive course · by Asem Abdo · for engineers who already use coding agents*

You already ship code with Claude Code, Codex or Cursor. This course gives you the discipline behind it: context engineering, harness engineering, loop engineering, orchestration and evals, in clear terms and with runnable examples. Every code sample is a small harness written in plain Python and was executed with the real interpreter when this page was built, so every output you see is real.

**The map**

## Four disciplines, one stack

The vocabulary moved fast. In February 2026 Andrej Karpathy proposed **agentic engineering** for professional work where you orchestrate agents instead of writing most code yourself. Under that umbrella, practitioners now use three more precise terms. They nest:

**Figure: Each layer contains the one inside it. The model reads only what the context window holds, so better context, harness and loops help more than better words..** Nested layers. Agentic engineering contains loop engineering, which contains harness engineering. Inside the harness, the context window holds instructions, tools, memory, history, tool results and the task, and the model reads the context window and proposes the next action.

> **Watch out:** **These terms are young.** Writers draw the boundaries differently: some treat harness engineering as part of context engineering, and "loop engineering" is still being defined. This course uses the layering above, explains each choice, and links every claim to its source on the Roadmap page.

**The three mental models**

## Everything here comes back to these

**Three rules that explain agentic engineering**

1. **Agent = model + harness.** The same model performs very differently depending on its instructions, tools, permissions and context. Most reliability comes from the system around the model.
2. **Enforce, don't request.** A rule in a prompt is a request the model may forget. A rule in a hook, a schema, a test or a permission is a guarantee.
3. **The context window is a budget.** Every token competes for attention and money. Load what the step needs, when it needs it, and nothing else.

**Short on time**

## The 90-minute route

1. **Vocabulary and the loop**: [Page 2](https://openplanr.dev/learn/agentic-engineering/gently) and [page 3](https://openplanr.dev/learn/agentic-engineering/agent-loop). 25 minutes.
1. **Context and harness**: [Page 4](https://openplanr.dev/learn/agentic-engineering/context-engineering) and [page 5](https://openplanr.dev/learn/agentic-engineering/harness-engineering), the core. 30 minutes.
1. **Loop engineering**: Goals, checks, stop conditions: [page 7](https://openplanr.dev/learn/agentic-engineering/loop-engineering). 15 minutes.
1. **Lab + cards**: L1 drills, then 20 flashcards. 20 minutes.

## Course outline

1. [Agentic engineering, gently](https://openplanr.dev/learn/agentic-engineering/gently.md)
2. [The agent loop and tools](https://openplanr.dev/learn/agentic-engineering/agent-loop.md)
3. [Context engineering](https://openplanr.dev/learn/agentic-engineering/context-engineering.md)
4. [Harness engineering](https://openplanr.dev/learn/agentic-engineering/harness-engineering.md)
5. [The primitives, hands on](https://openplanr.dev/learn/agentic-engineering/primitives.md)
6. [Loop engineering](https://openplanr.dev/learn/agentic-engineering/loop-engineering.md)
7. [Multi-agent orchestration](https://openplanr.dev/learn/agentic-engineering/multi-agent.md)
8. [Proving it: evals, cost, security](https://openplanr.dev/learn/agentic-engineering/evals-cost-security.md)
9. [Lab](https://openplanr.dev/learn/agentic-engineering/lab.md)
10. [Roadmap and sources](https://openplanr.dev/learn/agentic-engineering/roadmap.md)
11. [Glossary](https://openplanr.dev/learn/agentic-engineering/glossary.md)
12. [Flashcards](https://openplanr.dev/learn/agentic-engineering/flashcards.md)