§ 01
One-Page Snapshot
Workshop Overview
Learners move from "I use AI sometimes" to "I can design a simple system that captures work, routes it, adds AI judgment, and keeps a human in the loop."
Session Title
Build Your First AI Agent
What This Workshop Is About
Building useful no-code systems: first a simple automation, then an AI-enhanced workflow, then a narrow goal-driven agent.
Why It Matters
The biggest leverage comes from automating repetitive work, then adding AI where information needs to be summarized, classified, drafted, or evaluated.
Intended Audience
Beginners and working professionals who want practical automation skills without writing code.
Prerequisites
Comfort using web tools. Access to a form tool, Google Sheets, Slack or email, and Relay or a comparable no-code automation tool.
Running Example
A lead capture workflow: form submitted, lead stored, notification sent, AI summarizes and classifies, then an agent evaluates and drafts a response.
Recommended Timing
120 minutes total: concept framing, workflow mapping, live build, AI upgrade, agent design, and final share-out.
Final Output
A working first automation plus a scoped first AI agent with a goal, tools, instructions, boundaries, and review step.
Learning Outcomes
By the end of the workshop, participants will be able to:
- Identify tasks worth automating using repetition, low judgment, and friction as signals.
- Map a workflow in plain English before choosing tools.
- Build a simple Relay automation with one trigger and at least two actions.
- Add an AI step that summarizes, classifies, extracts, drafts, or transforms information.
- Explain the difference between a fixed automation and a goal-driven AI agent.
- Design a narrow AI agent with clear goals, tools, instructions, boundaries, and human review.
Core Thesis
Automation handles the structure. AI handles the flexible thinking inside the structure. Agents add a narrow goal and the ability to decide which steps to take.
Workshop Thesis
§ 02
Minute-by-Minute
Run of Show
Keep the build sequence linear. The same lead workflow should evolve across the full two hours so learners can see how automation becomes AI workflow, then agent.
| Time | Type | Topic / Activity | Facilitator Cues |
|---|---|---|---|
| 0:00-0:10 | Frame | Frame + Mental Model + Use Case Define automation, explain the automation to AI workflow to agent progression, and introduce the lead-capture use case. |
Make the workshop arc visible before opening tools. |
| 0:10-0:22 | Activity | Find Your Automation Target Learners list repetitive, low-judgment, annoying work and choose one target. |
Land the formula: Frequency x Time x Annoyance. |
| 0:22-0:38 | Map | Map the Lead Capture Workflow Map the shared use case using Trigger, Inputs, Steps, Output. |
Keep it plain-English and linear before touching tools. |
| 0:38-0:43 | Demo 1 | Basic Automation Show the Relay form trigger, Sheets row, and Slack/email notification. |
Five minutes. Show the target build end-to-end. |
| 0:43-1:05 | Practice 1 | Build & Test Learners build and test the basic Relay automation. |
Working automation or documented blocker. |
| 1:05-1:35 | Demo 2 + Practice 2 | Add AI Demo the AI step, then learners add summarization or classification to their workflow. |
Demo for five minutes, then move directly into practice. |
| 1:35-1:45 | Concept | Agents: What They Are & How to Scope Contrast fixed automations with goal-driven agents, then define goal, tools, instructions, boundaries, output, and review. |
Push against hype and protect scope. |
| 1:45-2:00 | Demo 3 + Practice 3 | Build the Agent + Close Demo the narrow agent, learners build or draft their first version, then close with humans in the loop, share-out, and takeaways. |
End with a review checkpoint and named next test. |
If You're Behind
Compress tool setup, not the workflow mapping. If time is short, demo the Relay build live and make the final agent a scoped design instead of a live build. Never skip the distinction between automation, AI workflow, and agent; that distinction is the conceptual spine of the workshop.
§ 03
Deck Standards
Slide Deck Guide
Slides should make the system visible: one workflow, getting more capable across the two-hour session.
Design Requirements
One running workflow diagram appears throughout
Start with Form submission → Sheet → Notification. Add AI summary/classification after the build lab. Add research, evaluation, and draft response during the final agent build.
Use real interface screenshots
Include screenshots of the form, Relay trigger, field mapping, Google Sheet row, notification, AI step configuration, and final agent output.
Keep terms plain-language
Every technical word gets a working definition: trigger, input, action, output, AI step, agent, tool, boundary, human-in-the-loop.
Speaker notes carry exact build steps
The slide should not be a manual. Put precise click paths, field names, and backup instructions in presenter notes.
§ 04
Contingency Plans
What Could Go Wrong
Most problems are caused by unclear workflow thinking, broken field mapping, or oversized agent scope.
Learners start in tools too early
Pause tool use and return to Trigger, Inputs, Steps, Output. Require a plain-English workflow before building.
Relay trigger does not fire
Check that the form was submitted after the automation was enabled. Run a fresh test and use the simplest possible form fields.
Fields map incorrectly
Reduce to name, email, and message. Verify each field one at a time before adding optional fields.
AI output is vague
Add criteria and output format. Do not ask the AI to "analyze"; tell it exactly what categories, fields, or draft structure to produce.
Agent scope gets too broad
Narrow the goal to one job. Good: "Evaluate inbound leads and draft replies." Bad: "Manage my sales pipeline."
Learner wants full autonomy
Move the first version to recommendation or draft mode. Human approval comes before sending, scheduling, deleting, purchasing, or changing records.
Tool access or login fails
Switch to the worksheet version. Learners can still map the workflow, write AI prompts, and scope the agent without live tool access.
§ 05Pre-Flight
Before the Session
Before the Session
- Review source transcripts end-to-end
- Prepare the lead-capture form
- Create a clean Google Sheet for demo leads
- Confirm Relay account access
- Confirm Slack or email notification access
- Build the demo automation once before teaching
- Prepare a test lead message for AI summarization
- Prepare screenshots for each critical step
- Print opportunity and workflow worksheets
- Print agent design canvas
- Prepare fallback if tool access fails