Andrew Yeung left Google and Meta and built multiple 7-figure businesses without a traditional team. His secret? An AI-augmented toolstack that automates coding, meeting notes, voice-to-text writing, and CRM management—all coordinated through a cloud-based “second computer.” This is the definitive breakdown of the 14 tools that replace an entire back office.
Andrew Yeung’s résumé reads like a tech whirlwind: former Google and Meta employee, host of 100+ annual events for 20,000+ people across New York, San Francisco, and Austin, angel investor in 20+ startups, and operator of two companies. Yet, he runs this multi-million-dollar ecosystem without a large team. The engine isn’t hustle—it’s a meticulously integrated stack of AI tools that function as a virtual chief of staff, database administrator, developer, and executive assistant.
Yeung’s approach treats AI not as a single app but as an operating system layer that connects every workflow. The result is a solo founder’s capability to manage events, media, investing, and advisory work at scale. Here is the exact stack, categorized by function, with analysis of how each piece fits into a system that replaces traditional headcount.
AI as an Operating System: Claude Code and Zo Computer
The foundation of Yeung’s stack is two “AI computers” that handle thinking and execution.
Claude Code is his primary LLM interface, but it’s treated as a chief of staff, not just a chatbot. Through a Model Context Protocol (MCP) integration, Claude Code connects directly to his Notion workspace, allowing it to read, write, and organize project management data without manual prompting. It also plugs into his Customer Relationship Management (CRM) system, drafts content, stress-tests business ideas, and writes emails. The key insight: context persistence. Claude doesn’t start from scratch each session; it operates within the ongoing context of his businesses as reported by Business Insider.
Zo Computer is a physical server he’s never touched—a cloud-based AI computer that runs while he sleeps. It hosts lightweight custom tools, runs background monitoring, and automates tasks that previously required either a dedicated assistant or significant manual overhead. This is the “second computer” that handles asynchronous work, effectively giving him two operational brains: one for real-time interaction (Claude) and one for background compute (Zo).
The Data Layer: Notion and Airtable
Yeung separates “thinking” from “tracking.”
Notion serves as the primary database for all content creation: TikTok and LinkedIn scripts, drafts, calendars, and status tracking for every deliverable. Its power comes from the MCP connection to Claude Code, which automatically categorizes, schedules, and organizes content without manual updates [see Business Insider’s coverage of Notion’s role in creator workflows].
Airtable handles structured operational data: contact forms, event registrations, and partnership inquiries. It’s more relational than Notion, built for querying and acting on data. Every time someone registers for an event or applies to his hiring series, that data flows into Airtable automatically, creating a clean, actionable dataset for follow-up and analysis.
Rapid Prototyping and Build: Lovable and Framer
The ability to build custom software without engineers is a game-changer for solo founders.
Lovable is a no-code prototyping tool that Yeung used to build a custom matching algorithm for his hiring event series. What would have taken months and a significant budget was built in a weekend. He describes the workflow: “I’m not an engineer, but I can describe a problem, iterate on a solution, and ship something functional in hours.” This collapses the development lifecycle from concept to deployment [Business Insider profile on Lovable’s CEO].
Framer is his web design platform, but he uses it as a full brand-building tool. It allows design-first site creation without developer dependencies, meaning he can update landing pages for events the night before they happen. This agility is critical for a business with constantly changing event schedules and partnership pages.
Meeting Memory and Voice-First Work: Granola and Wispr Flow
Cognitive overhead is the silent killer of multi-business founders. Yeung’s stack aggressively captures and automates communication.
Granola is an AI note-taking app that transcribes, summarizes, and analyzes meetings in real time. Its killer feature: after a meeting ends, it uses the transcript to draft accurate, personalized follow-up emails automatically. For Yeung, this means sponsor conversations, investor check-ins, and partnership negotiations are all summarized, searchable, and actioned without manual note-taking.
Wispr Flow turns his microphone into a primary input device. It’s a voice-to-text tool that works across every application, allegedly four times faster than typing with built-in AI commands and auto-edits. Yeung uses it to draft emails, LinkedIn posts, and partnership briefs. His first drafts are now almost entirely spoken, dramatically reducing the friction of writing.
Communication and Context Management: Kondo and Raycast
When your business happens in LinkedIn DMs and text messages, inbox zero isn’t a luxury—it’s revenue-critical.
Kondo is dubbed “Superhuman for LinkedIn.” It labels conversations, snoozes follow-ups, processes messages in bulk with keyboard shortcuts, and ensures no business opportunity slips through. For a founder who generates partnerships and sponsorships via LinkedIn, this is the single biggest workflow upgrade he made this year, directly turning messaging efficiency into deal flow [Business Insider on managing high-volume comms].
Raycast is a keyboard-first launcher that eliminates mouse-driven context switching. From a single shortcut, he can jump between tools, insert saved text snippets for repeated messages, launch automations, and even trigger Claude AI from any screen. This might seem minor, but in a day of 50+ context switches across multiple businesses, shaving seconds each time compounds into hours saved.
Audience and Publication: Substack and Beehiiv
Yeung runs two newsletters for two distinct audiences, using two specialized platforms.
Substack powers his personal blog and newsletter for his thoughts on entrepreneurship, events, and sobriety. He chooses it for its clean writing environment and its built-in community of serious writers, which aligns with his personal brand.
Beehiiv powers the newsletter for The Shortlist NYC, his hiring events series. Beehiiv is chosen for its growth-oriented analytics and infrastructure suited for a business publication, not just a personal blog [Business Insider on Beehiiv’s business model].
State Management: Endel and Othership
None of the tools matter if the operator’s cognitive state is degraded. Yeung explicitly invests in tools that manage his focus and recovery.
Endel generates adaptive soundscapes calibrated to focus, sleep, and recovery. He runs it in the background during deep work sessions, claiming it keeps him “in the zone” better than static playlists.
Othership is a guided breathwork and meditation app he uses as a deliberate “off-ramp” at day’s end. In a work life with no clear boundaries—always the founder, creator, and investor—this is non-negotiable maintenance.
The Integration Stack That Makes It All Work
Yeung’s system isn’t a random collection of apps; it’s an integrated pipeline:
- Data Ingest: Event registrations and leads flow into Airtable automatically.
- AI Processing: Claude Code, connected via MCP to Notion, organizes and prioritizes this data without manual entry.
- Background Compute: Zo Computer runs monitoring and lightweight custom tools (like the matching algorithm from Lovable) around the clock.
- Communication Output: Granola drafts follow-ups; Wispr Flow powers first drafts; Kondo manages LinkedIn replies; Raycast speeds up everything.
- Audience Publication: Substack and Beehiiv distribute content to segmented audiences.
The connective tissue is MCP integrations and cloud-based computing. This is the modern solo founder’s stack: replace administrative, developmental, and communicative labor with AI-augmented tools that talk to each other.
Why This Matters for Developers and Founders
Yeung’s stack is a blueprint for building a capital-efficient, high-output business in 2024–2025. The implications are immediate:
- For Solo Founders: You can now run what used to require a COO, a developer, a marketing manager, and an executive assistant. The barrier to scaling a service or events business has plummeted.
- For Tool Builders: The winning products are those that offer deep integrations (like Claude’s MCP support) and act as platforms, not point solutions. Tools that don’t connect will be isolated and eventually replaced.
- For Investors: A founder’s ability to articulate their toolstack is now a due diligence item. It reveals operational sophistication and capital efficiency. A team of five using this stack might outperform a team of twenty using siloed tools.
- For Enterprise: This is the future of internal productivity. The same tools—cloud computers, AI note-takers, voice interfaces—will soon be deployed within large organizations to reduce managerial overhead.
The takeaway: scalability is no longer tied directly to headcount. It’s tied to system design. Yeung’s 7-figure career with a team of one proves that the “solopreneur” ceiling has been shattered—not by working harder, but by wiring together the right AI-augmented tools.
This analysis is based on the firsthand account of Andrew Yeung as reported by Business Insider and corroborated by product-specific coverage of each tool’s capabilities within the tech ecosystem.
For the fastest, most authoritative breakdowns of how AI is reshaping entrepreneurship, onlytrustedinfo.com delivers actionable insights without fluff. Our technology desk tracks the toolstacks and systems that actually move the needle—so you can build smarter, not harder. Read more analysis of the AI-driven solopreneur revolution in our AI Tools section.
