Agentic Engineering - Delegate Real Work to AI Agents
Agentic Engineering: Delegate Real Work to AI Agents
https://WebToolTip.com
Published 10/2026
Created by daisuke matsutoya
MP4 | Video: h264, 1920x1080 | Audio: AAC, 44.1 KHz, 2 Ch
Level: Intermediate | Genre: eLearning | Language: English | Duration: 6 Lectures ( 2h 46m ) | Size: 2.1 GB
Spec writing, review, evaluation loops, and directing more than one AI agent at once -- hands-on, no coding required.
What you'll learn
⚡ Explain what an AI agent actually is (reasoning core, tools, loop) and diagnose where your own AI use sits on the 4-stage adoption ladder
⚡ Write a three-part goal packet (goal, done-condition, boundaries) that an agent can run with unsupervised
⚡ Apply the 4 governing principles behind a reliable instruction sheet (think before acting, keep it simple, make changes surgically, drive from the goal)
⚡ Review an agent's output for "looks right but is wrong" mistakes, using a structured review checklist
⚡ Build a self-verification loop so an agent checks its own work before reporting back, without you being in the loop every cycle
⚡ Direct more than one agent at once using role separation -- a builder, a critic, and a judge
⚡ Design a multi-agent pipeline using the core patterns: separate deterministic from non-deterministic steps, add checkpoints, start small
⚡ Plan a 3-stage rollout of agent delegation in a team or company, and pick the two numbers worth measuring as you scale
Requirements
❗ A computer (Mac or PC) is required.
❗ A Claude account (or equivalent AI agent tool) is required.
❗ No programming experience is required -- every technique in this course is a written instruction and a review process.
❗ Prior comfort with an AI chat or agent tool is recommended (this course reviews the basics briefly but isn't a from-zero introduction).