Andrew Wilkinson: AI Agents run my business and life

May 14, 2026

AI Summary

5 min read

Andrew Wilkinson describes how he has integrated AI agents into both his operating companies and personal systems, using tools such as OpenClaw and a custom harness called Harbor to handle routine execution while he focuses on higher-level direction. The approach treats agents as capable but immature assistants that require explicit process definitions and ongoing oversight rather than operating as fully independent operators.

Agent Setup for Business Operations

Wilkinson tested the model by launching a small SaaS product called Deep Personality entirely through agents. The system routes support tickets to either a dedicated support agent or a development agent depending on severity, with the development agent able to generate and merge code fixes in straightforward cases. A marketing agent connected to usage data runs multivariate tests, generates creative, and adjusts budgets on Meta and Reddit, subject to periodic human approval on larger spends. He reports that these loops now produce revenue in the low tens of thousands of dollars per month, though the largest share of his time remains split between debugging agent behavior and refining instructions. Current agents function more like scripted workflows with decision branches than autonomous managers; they still need detailed step-by-step guidance on cadence, escalation, and output formats.

Continue reading the full summary in the app — free to try.

Read Full Summary →

Free • No credit card required

Listen to Audio Summary Open in App

What you'll learn

1 (01:50) **Andrew's shift to AI agents** - Describes discovering Claude Code in late 2025 and becoming obsessed with building agents
2 (03:34) **Autonomous SaaS test business** - Built and ran a full SaaS company using OpenClaw agents
3 (05:09) **Deep Personality app creation** - Built a 40-minute personality test and report generator in days
4 (10:30) **Agent orchestration with Harbor** - Replaced OpenClaw with Harbor for reliable multi-agent system
5 (12:39) **Current limits of autonomous companies** - Explains why fully hands-off AI companies are not yet real
6 (17:31) **Credibility challenge for AI products** - Discusses need for expert validation on Deep Personality
7 (20:14) **Centralized knowledge with vector databases** - Built searchable company memory using GBrain and Pinecone

+ Full timestamped outline available in the app

Show Notes

If you want more workflows and tactics to build a business with AI, check out this free workshop: https://www.ideabrowser.com/workshop

I sit down with Andrew Wilkinson and we go deep on how he's restructured his work, his health, and his family office around AI agents. Andrew walks me through Deep Personality (an app he vibe-coded after running psychological screens on himself and his girlfriend), the autonomous SaaS business he runs through agent harnesses like Harbor, and the vector-database setup that lets him query Tiny and his personal holding company like an oracle. We cover where software is headed, why he's pouring capital into TSMC and data center stocks, and the daily AI workflows he's built around health, email triage, and a personalized morning podcast. Listeners walk away with concrete prompting tactics, agent architectures, and a frank read on where the moats are moving.

Timestamps:

00:00 – Intro

01:50 – The OpenClaw and Claude Code Unlock

04:53 – Demo: Deep Personality App

10:38 – Harbor: An Agent Harness For Real Companies

12:30 – Autonomous Companies: Hype Vs. Reality

17:30 – Credibility As The Missing Layer For Vibe-Coded Products

20:14 – Centralizing Data Pipelines

21:35 – Vector Databases

23:22 – Transitioning Companies to Agentic Companies

25:22 – Where Andrew Would Build Today

27:10 – The New Interface

28:21 – Why build now

30:59 – Replacing Adapar: A Networth Wealth Platform

33:07 – Services As The New Software

35:24 – G-Brain Explained and Andrew’s OpenClaws

45:09 – Closing Thoughts

Key Points

Andrew runs a SaaS business called Deep Personality almost entirely through agents, generating roughly $20K of revenue while debugging eats half his time.
Harbor (github.com/geekforbrains/Harbor) gives agents a GUI-style harness — dev, marketing, and support agents that can autonomously merge PRs and adjust ad budgets across PostHog, Meta, and Reddit.
Andrew's family office swapped headcount for a $40K/month Claude bill; his CFO, who had zero coding background, vibe-coded a replacement for Adapar (priced at $50K–$100K/year) in about two weeks.
Vector databases trained on Tiny and Andrew's holding company let him query 132 minority i