[REDACTED] Episode 8: We Hired an AI Sales Rep and Became Its Manager
On Episode 8 of Redacted, we walk through the daily AI agent loop that now runs our sales pipeline and the real math on how many hours AI-assisted work actually saves.
Episode 7
Redacted is the show that doesn't clean things up before hitting record. Episode 8 is almost entirely show-and-tell: Taylor demos the agent system now running Offline's lead gen and HubSpot cleanup on autopilot, and David walks through a redesigned mobile app, a new email system, and a surprisingly candid hour-by-hour accounting of what AI-assisted work costs versus the old way of doing it. This is our biweekly, screen-share-first look at what it's actually like to build a company with AI in 2026 with no influencer takes and no model hype.
What We Cover
The “outreach brief” loop: Every morning at 6am, Taylor’s agent scans market signals, matches them to companies in HubSpot (creating new records when needed), and produces a concise outreach brief with everything Offline’s sales lead needs before contacting a restaurant.
“Now I’m the manager of this employee”: David describes Taylor’s role as managing the AI system, reviewing its overnight work, resolving escalations, and improving Claude Code “skills” so recurring mistakes don’t happen again.
CRM drift as a law of physics: David argues that any two systems kept in sync will inevitably drift apart, which is why they built a daily AI agent to detect and reconcile differences between Offline’s backend and HubSpot.
The jungle-and-bulldozer metaphor: Steve, the sales lead, uses human judgment to carve the initial path through the “jungle,” while builders automate and scale that proven path into a repeatable process.
Structured feedback beats meetings: Rather than discussing outreach briefs in meetings, Steve leaves comments directly in HubSpot, allowing the AI to categorize feedback into structured data instead of relying on verbal discussions.
The silent email failure: Taylor shipped an AI-built product announcement email with a broken unsubscribe link, causing 10,000 emails to fail compliance checks without anyone noticing for seven days. The incident reinforced the need for careful review.
64.39 hours, broken down: David spent 16 hours refining positioning, 6 hours on visual design, and 42 hours building the landing page. He believes this took far less time than aligning stakeholders through traditional meetings.
74 turns with Claude: It took 74 back-and-forth exchanges with Claude to refine the landing page, illustrating that effective AI-assisted design still requires substantial iteration rather than one-shot prompting.
Compounding as the real unlock: After investing time to build the initial design system, David estimates recreating a similar page would take around 30 hours. Adapting it to mobile is now largely a matter of prompting.
The daily biz dev brief: David’s AI monitors more than 20 restaurant technology companies and surfaced timely industry developments, including Capital One Dining shutting down, reports of Amex pursuing TheFork, and intensifying competition among OpenTable, SevenRooms, and DoorDash.
The “hiring” framing: David compares AI agents to new employees, arguing that intelligence alone is not enough. They require onboarding, training, clear processes, and ongoing supervision to perform reliably.
Two builders, one company, and a running tally of what AI actually costs and saves in hours. No hype, just the receipts. Enjoy the conversation.
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.