[REDACTED] Episode 7: How a Marketer with One Broken Assembly Class Built a Lead-Gen Tool Solo
Ben Pope joins Redacted to show how he vibe-coded a lead-scoring web app from scratch and how his new AI Solutions unit is turning client consulting into product roadmap.
Episode 7
Redacted is the show that doesn't clean things up before hitting record. Redacted's first-ever guest is Ben Pope, who spent five years in marketing at Levitate before pivoting a month ago to lead the company's brand-new AI Solutions business unit. This one leans heavy on show-and-tell: Ben screen-shares a lead-scoring tool he built solo with Claude, a morning-briefing skill that preps him for every meeting, and a post-demo skill that drafts follow-ups and updates the CRM automatically, and then the three of us dig into the founder-lesson stuff, like when to trust an LLM versus a deterministic script.
Who the Guest Is
Ben Pope has spent the last five years at Levitate, joining on the marketing side when the company was doing about $6 million in ARR. It’s since grown to roughly $45-50 million. Levitate builds software and services for small businesses (independent insurance agents, RIAs, attorneys, nonprofits, home services) to help them stay in touch with their networks and drive referral and word-of-mouth revenue, the biggest growth lever for most small businesses. A month before this episode, Ben stepped away from marketing to launch and lead Levitate’s new AI Solutions business unit, helping small-business clients figure out how to fold AI into their own workflows. He describes himself as a self-taught tinkerer rather than an engineer since he took a couple of computer science classes in school but says an assembly course “broke him” and he taught himself to build and ship full web apps using Claude, GitHub, Vercel, and Neon.
What We Cover
From $6M to ~$50M ARR: Ben joined Levitate five years ago when it was doing about $6 million in ARR; the company has since grown to roughly $45–50 million in ARR, serving nearly 9,000 SMB customers with 300+ employees.
The founding story: Levitate’s thesis traces back to CEO Jess’s wife, Brooks Bell, whose marketing consultancy ran almost entirely on word-of-mouth. The software she bought to manage it was too complex to self-serve, inspiring the idea for a software-plus-service model built specifically for small businesses.
“I can throw Claude at that”: Ben’s litmus test for adopting a tool is simple. If it requires clicking through settings or figuring out an API, he hands the task to Claude instead.
Solo-built, no engineering background: Despite describing himself as “not really that sharp” with code and saying a college assembly class “broke him,” Ben built and deployed a full lead-scoring web app using GitHub, Vercel, and Neon.
Killing Webflow entirely: Levitate is migrating its marketing site from Webflow to a GitHub-and-Claude-managed workflow so anyone on the team can update the site by describing changes to Claude instead of learning Webflow.
Trained the CRO into a Claude convert: Ben personally taught Levitate’s CRO how to use Claude, leading him to stay up until 2–3 a.m. automating workflows and eventually becoming an internal AI evangelist.
Deterministic vs. probabilistic: Ben’s rule is that financial reporting and other “sacred” data should always be generated with deterministic scripts first, with LLMs layered on only where necessary to minimize hallucination risk.
~100 skills built, most abandoned: Ben estimates he has created around 100 Claude skills over the past year, many of which were abandoned after breaking. His solution is to periodically have Claude audit which skills are still actively used.
AI Solutions as a forward-deployed engineering arm: Consulting projects, such as intake automation for an LA family-law attorney, serve as R&D. Successful patterns are incorporated into Levitate’s core product, including a meeting agent feature that originated from client work.
Recommended resource: Ben recommends Simon Willison’s blog as essential reading for agentic engineering patterns and AI-focused test-driven development, and says his entire team follows it.
It’s best viewed on YouTube to fully see the examples (make sure to subscribe!)
But also available on all audio podcast players through Tweener Talks!
PLUS we have a new spot for show notes and files discussed in the episode. Check it out: https://github.com/instanttaylor/redacted-podcast
We need your [Redacted] AI experiences for upcoming episodes! Who’s the most underrated AI builder you know? Someone running real systems inside a real business? Send us a message at contact@tweenerfund.com because we want to get them on the show!
What’s Next?
New episodes drop twice a month/every other Wednesday. If you want to be on the show as a guest and show your [REDACTED] builds, email us here.