The Simplest Way to Build an AI Assistant That Actually Works with Jeff Ignacio
Agentic Workflows Should Be Hard or Take Months to Learn – This No-Code Tool Simplifies It
I’ve known Jeff Ignacio for years from the RevOps world. He’s one of those rare operators who actually builds the things he talks about. So when I saw he was deep in the AI trenches, wiring up no-code agents in Zapier, I knew we had to have him on.
In this episode, Jeff walks me through how he built his own executive assistant agent, one that reads meeting notes, drafts follow-ups, and drops a daily checklist in Slack before he’s even had his coffee. It’s practical, accessible, and something you can build today.
This week’s Stack:
1 video: Build a no-code executive assistant agent in Zapier
1 prompt: Design your agent’s “job description” for reliability
1 tool: Route each content task to the right model
3 resources: Smart workflows, better prompts, fresh trends
3 jobs: Killer marketing roles at top AI companies
Let’s go!
Workflow Walkthrough: Build a No-Code Executive Assistant Agent in Zapier
This week, Jeff Ignacio joined me to demo something deceptively simple, an AI agent that actually works.
He wired Gmail, Calendar, Zoom, Slack, and Asana together in Zapier to build a personal “executive assistant” that reviews meetings, drafts follow-ups, and sends him a daily checklist in Slack.
Here’s what you’ll learn:
How to configure a Zapier agent to read Otter summaries, scan Gmail, and compile action items.
Why to use “create draft” guardrails in Gmail and Slack as your notification surface.
How to give your agent a clear role, context, and markup guide for clean outputs.
When to keep a human in the loop and how to think about failure modes.
Jeff’s favorite AI stack: Zapier Agents, Clay, Claude, and more.
Prompt of the Week: Strategic Media Campaign Builder
Everyone wants a quick media campaign, but no one’s willing to put in the work to set themselves up for success. Here’s a repeatable path to campaigns that feel fresh and actually move pipeline. So, the next time your boss tell you "just make content” you’ll have this handy prompt that acts as your strategic creative director: it collects the product, profiles the audience, shapes topics, and holds the final concept until you’re ready to review.
SYSTEM MESSAGE:
You are a world-class marketing strategist specializing in differentiated media campaigns. Your task is to guide the user through a structured, multi-step process to design a fun, engaging, and effective media strategy for their product. Your tone should be creative, energetic, and insightful — while always following each step exactly as outlined. Do not output final campaign results until explicitly asked. Store all intermediate findings in variables for later reference.
USER PROMPT:
Please perform the following task in sequence. Output only when the final step is reached and the user has explicitly requested the output.
TASK:
Create a fun, engaging, and differentiated media strategy for a product’s target audience.
INPUTS:
$Product_Name — to be collected from the user at the beginning.
STEMPS (Complete in Order):
1. Collect Product Name
Ask the user: “What is the name of the product this media strategy is for?”
Store in $Product_Name.
2. Audience Analysis
Research and capture the following:
- $Persona_Overview: A short summary of the typical buyer — their job, the problem they face, and how the product solves it.
- $Persona_Topics: Key content topics the buyer persona is interested in that align with the product. These topics should be relevant to potential purchase intent.
- $Persona_Mediums: Core content formats this persona prefers (e.g., blogs, podcasts, YouTube, TikTok, newsletters).
3. Content Strategy
Using the above audience insights, generate a list of content topics this product should create content around to attract and convert the target audience.
Store in $Content_Topics.
4. Media Campaign Design
Based on all prior variables ($Persona_Overview, $Persona_Topics, $Persona_Mediums, $Content_Topics), do the following:
- Select one content medium from $Persona_Mediums that is the most promising.
- Design a creative and engaging media campaign using that medium. Describe:
- The campaign concept and how it fits the persona’s interests.
- Why it’s fun and engaging.
- Potential creators/influencers to partner with (and how to uniquely integrate them).
- Include a section titled: “How to 10x This Idea” — describing how to dramatically enhance the campaign’s impact.
IMPORTANT:
- Execute each step in order.
- Do not skip or summarize.
- Wait for the user to request the final campaign output before revealing step 4.I’ve added this to my The Ultimate ChatGPT Prompt Library for B2B Marketing Leaders notion doc. Check it out for 60+ more prompts.
The Ultimate ChatGPT Prompt Library for Marketing Leaders
I remember the first time I used ChatGPT for marketing. It was late, I was up against a deadline, and I needed a competitive analysis that would have taken me and my PMM a full day (or more) to pull …
Know someone who might find this prompt useful? Share it with them!
Tool of the Week:
AI Resource Roundup:
How AI Reshapes the Revenue Operations Function (RevOps Impact): Jeff Ignacio breaks down how AI is transforming RevOps, from forecasting and data enrichment to automation and decision support. It’s a grounded look at how operations leaders can stay strategic as the tech shifts under them.
SaaStr’s Top 10+ AI Agents (YouTube): SaaStr reveals how they went from zero to twenty AI agents in nine months. You’ll see their internal “AI SDRs,” “AI BDRs,” and “AI RevOps” systems in action, all built with Replit and off-the-shelf tools. It’s a real-world playbook for scaling agentic operations.
Prompt Optimizer from OpenAI: A simple but powerful tool inside the OpenAI platform that helps refine prompts to best-practice standards before use. Paste any prompt, and it returns an optimized version ready for production.
Here’s a quick video of me testing it out and comparing it to a custom GPT I built using OpenAI’s prompting guide and documentation.
Hot AI Jobs: The Week’s Most Interesting Open Roles
Director of AI GTM at MaintainX
Location: Remote (North America)
Pay: Not listed
Product Marketing Manager at Recall.ai
Location: San Francisco, CA (In-Person)
Pay: ≈ $181,400 / year
Senior Product Marketing Manager at Airwallex
Location: San Francisco, CA (Hybrid)
Pay: $175,000 – $215,000 / year
I’ve been in a building mood lately… maybe it’s the fall weather, maybe it’s the flood of new AI tools. Either way, every conversation lately seems to circle back to one thing: reliability. Jeff’s no-code assistant nailed that theme, small scope, big impact, human in the loop.
If you try building your own version in Zapier, tag me. I’d love to see what you come up with!
– Brandon
 | A guest post by
|