Jun 26, 2026 · 33m · saastr
How to Build Your Own AI VP of Marketing Step-by-Step
0:00 / 0:00
▶ Watch on YouTube →
gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions
In this hands-on workshop, the presenter demonstrates how to build and deploy an autonomous AI VP of Marketing on Replit using structured specifications, sample datasets, and API integrations. Modeled after SaaStr's internal AI agent Tenke, the session provides a comprehensive blueprint for automating marketing workflows, analyzing pipeline data, and implementing essential human-in-the-loop guardrails.
How this conversation actually went
Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. How this is scored →
Jason as informed peer 0.0 Guest teaching 0.0 Guest disagreement 0.0 Jason pushing back 0.0
speaking balance: gold is Jason, purple is the guest (3 minute bins)
Sharpest disagreement ▶ 26:40 Calling out sales reps for unkept network promises
In a humorous critique rather than interpersonal hostility, Amelia notes how Account Executives claim they will work their Rolodex but fail to follow through without agent automation.
Hardest push from Jason ▶ 0:41 Monologue format baselineThere is no host pushback in this episode because the recording is a solo workshop presentation delivered entirely by Amelia LeRutte.
Biggest teaching moment ▶ 28:00 Catching data hallucinations in campaign specsAmelia educates the audience on agent pitfalls, sharing an example where Tenke fabricated VC invite lists until forced to verify against real database records.
Jason holds their own ▶ 0:41 Solo presentation baselineThe episode is a solo presentation by Amelia without active host dialogue or host-driven technical corrections.
the scores for every segment, with the reasoning behind each
| Chapter | Topic | Jason as informed peer | Guest teaching | Guest disagreement | Jason pushing back | Why |
|---|---|---|---|---|---|---|
| Workshop Kickoff and the Origin of Tenke | 0 | 0 | 0 | 0 | Amelia opens the live workshop introducing Tenke, an internal AI agent created to automate tedious Sunday night reporting. As this is a solo workshop presentation without host participation, host scores and combativeness are zero. | |
| The Co-Pilot Mindset and Vibe Coding on Replit | 0 | 0 | 0 | 0 | Amelia explains the concept of treating AI agents as distinct copilot personalities rather than simple chatbots and introduces Replit vibe coding. The format remains a pure solo instructional presentation. | |
| Defining the Spec, Data Foundations, and Agent Functions | 0 | 0 | 0 | 0 | Amelia breaks down how to construct a functional spec, detailing CRM connections and focusing the agent on a single primary goal like pipeline revenue. There is no host involvement or disagreement. | |
| Initiating the Live Build with Specs and Sample Datasets | 0 | 0 | 0 | 0 | Amelia instructs the audience on downloading specs and sample datasets, initiating a live build prompt inside Replit while walking through the necessary data foundations. | |
| Reviewing Pre-Built Outputs: Dashboards, Outreach, and Daily Ideas | 0 | 0 | 0 | 0 | Amelia showcases pre-built outputs from Tenke, including win-back outreach lists and daily marketing suggestions, demonstrating how fine-tuning improves idea quality over time. | |
| Connecting APIs, Salesforce Apps, and Automating Menial Tasks | 0 | 0 | 0 | 0 | Amelia outlines connecting Salesforce Connected Apps and Google Calendar APIs, highlighting how autonomous calendar invites saved weeks of manual operational work. | |
| Automating Complex Campaigns: Triggered Emails and Newsletters | 0 | 0 | 0 | 0 | Amelia details advanced campaign automation including triggered competitive follow-ups and building a dedicated newsletter builder directly inside the agent environment. | |
| Development Best Practices: Stair-Stepping and Direct Interaction | 0 | 0 | 0 | 0 | Amelia advises incremental development ('stair-stepping') and directly interacting with the agent in the backend builder to refine its ad generation capabilities. | |
| Inspecting the Live Build and Tenke's Strategic Takeaways | 0 | 0 | 0 | 0 | Amelia reviews the finished live build, emphasizing guardrails, human-in-the-loop verification, and managing hallucinations when dealing with CRM data. |
Statements from this episode (12)
AI agents excel at ad creative due to endless ideation capacity
“Agents are really good at ads because they have, like, endless ideas, whereas I feel like as humans, if you want your ROAS to be really high, you need to change it, like, every other day or so and change it, like, seven times over.”
Interacting with an agent builder develops its distinct personality
“If you start to talk to the agent in this builder in the back end, it will develop its own, what I call personality or you can call it brain, but it will develop its own personality to do it.”
SaaStr operates nearly 30 AI agents used almost 1 million times
“We have, I don't know, almost 30 agents now at Saskr that have been used almost a million times.”
SaaStr built its AI agent over five months of daily work
“We've been building with Tenk literally every day between myself and Jason for the last five months, and so it might take you some time after this session, if you leave with a basic one, to build it to the full feature and functionality that we have at Saster.”
SaaStr's AI agent autonomously makes marketing decisions and executes campaigns
“So what 10 K is now, which I'll show you today, where he is making some autonomous marketing decisions. He helps us send campaigns. He's literally our co-pilot.”
AI marketing agents generate ideas exclusively grounded in data
“The nice thing about building something like an agent is any ideas it's going to give you is only based in data. Like, that's all it has. It's got all your data. It's got your numbers. It's got a goal, hopefully. You give it a goal, maybe write down your goal …”
AI agents must each be assigned a single, specific goal
“Make sure you give your agent one goal. So that's also why we have like multiple agents between our AI VP of Marketing, our AI VP of Customer Success, our AI Sash Reannual agent. Like, they all have different brains. I don't like to overload them. I like to ke…”
Semi-autonomous AI marketing agents require native CRM and MAP integrations
“Hook up natively To your marketing automation platform via the APIs hook up to your CRM, whatever you're using, some of your major tools. Again, you don't have to add those all day one. Eventually, you will need to, though, if you want to reach the same level …”
Revenue impact data is the primary dataset for marketing AI agents
“If you have any sort of data around marketing's final goal on an impact on revenue, that's the very first thing you're going to want to give your agent.”
Autonomous AI email reminders drove spikes in SaaStr ticket sales
“Again, this one collected emails when people put in for the code, so it already had their email, and then just started emailing them, I think only twice, autonomously on its own, and then every time we saw it send out daily a reminder, we would see ticket spik…”