Screensharing How This Guy Makes Millions with AI Agents

Key Concepts

• AI Agents for Business Scaling
• AI Avatar for Content Creation
• Voice-Dictated Writing for Content
• AI-Powered Meeting Scheduling
• AI Tutor for Learning from Videos
• GPT for New Hire Onboarding
• AI-Powered Email Filtering and Response

1. AI Avatar for Short-Form Video Content

• Problem: Founder Rowan Chung lacked time for creating short-form video content despite having written content in the newsletter.
• Solution: Cloned face and voice using AI to create an avatar that narrates short-form videos.
• Tools: HeyGen (avatar), 11 Labs (voice). Voice is generated separately in 11 Labs for higher quality and then stitched together.
• Process:
  1. Editor takes original writing and adapts it for short-form video using CloudScript.
  2. The script is fed to the AI avatar.
  3. Manual editing, storytelling, and B-roll are added.
• Results: Grew to over 150,000 followers in under a year.
• Key Argument: Effective for founders who are writers but lack time for video creation.
• Caveats: Requires a talented editor and isn't a "one-click" solution. Training the avatar is essential.
• Example: Viral video about Anguilla and .ai domains.
• Quote: Sam Altman's statement that it is the era of the idea guy.
• Counter-Argument: Using AI avatars may not be suitable for all situations, especially for public company CEOs who need to be authentic.
• Data: The avatar is only shown for the first few seconds of the video because the audio is more advanced than the video.
• Real-World Application: Faceless YouTube channels and animated characters demonstrate the potential for building a following through non-human content.

2. Voice-Dictated Writing for Tweets

• Problem: Need to create a lot of content quickly.
• Solution: Voice-dictated writing using AI trained on the user's writing style.
• Process:
  1. Train Claude with top-performing tweets using a specific prompt.
  2. Go on walks and dictate ideas using Whisper Flow.
  3. Push notes into Claude, which generates tweets.
  4. Edit and add human insight to the generated tweets.
• Prompt Example: Train on writing style by providing top-performing tweets.
• Results: Tweets are 90% ready, saving time and effort.
• Key Argument: Efficient way to generate a high volume of content.
• Example: Rowan goes on walks every Sunday to generate tweets for the week.
• Insight: The last 10% (adding human insight) is crucial for making the content engaging.
• Technical Term: Whisper Flow (voice-to-text app).

3. Using Claude as an Editor-in-Chief for Newsletter

• Problem: Need to refine written content for newsletters.
• Solution: Use Claude as an editor-in-chief to improve writing quality.
• Process:
  1. Write the newsletter content by hand.
  2. Paste the content into Claude and ask it to "make it better."
  3. Review Claude's suggestions and incorporate them into the original content.
• Key Argument: Improves content quality by leveraging AI for editing and refinement.
• Technical Term: Claude (AI assistant).
• Hack: When the chat gets too long, start a new chat and copy a small portion of the previous chat to maintain context.
• Example: Editing a main paragraph of the newsletter by asking Claude to refine it.
• Insight: It takes more time to edit with AI, but the content gets better.

4. Lindy AI Agent for Meeting Scheduling (Ava)

• Problem: Time-consuming process of scheduling meetings.
• Solution: Use Lindy to create an AI agent (Ava) that automatically schedules meetings.
• Process:
  1. CC Ava in an email thread.
  2. Ava looks at the user's Google Calendar and suggests available times.
  3. The other person responds, and Ava books the meeting.
  4. Ava can also reschedule meetings.
• Benefits:
  • Saves time and effort.
  • Responds to leads instantaneously (within a minute).
  • Works 24/7.
  • Avoids the stigma of sending a Calendly link.
• Customization:
  • Change the agent's name.
  • Connect your own email and Google Calendar.
  • Set meeting hours.
• Template: Lindy provides a template that can be modified.
• Real-World Application: Can be used for partnerships and sales.
• Example: Nicole, a lead, gets a meeting booked within minutes using Ava.

5. Perplexity Comet as an AI Tutor for YouTube Videos

• Problem: Difficult to quickly understand the key points of long YouTube videos.
• Solution: Use Perplexity Comet to summarize and learn from YouTube videos.
• Process:
  1. Install the Perplexity Comet widget.
  2. Open a YouTube video.
  3. Use Comet to ask questions about the video and get summaries.
• Benefits:
  • Summarizes long videos quickly.
  • Helps understand topics in more depth.
  • Turns every video into an AI tutor.
• Key Argument: Improves learning and retention from video content.
• Technical Term: Perplexity Comet (AI-powered search and summarization tool).
• Insight: Comet is grounded in the entire context of the video, making it smarter than a regular Perplexity search.

6. GPT for New Hire Onboarding

• Problem: Onboarding new hires is time-consuming and tedious.
• Solution: Create a GPT trained on company SOPs, Loom transcripts, meeting notes, policies, and HR docs.
• Process:
  1. Train the GPT with relevant company information.
  2. New hires ask the GPT questions about the company.
  3. If the GPT doesn't know the answer, it directs the user to the right person on the team.
• Benefits:
  • Saves time for managers.
  • Answers questions immediately.
  • Ensures consistent information is provided to new hires.
• Key Argument: Scales onboarding process and improves efficiency.
• Prompt Example: Create a custom onboarding assistant and drop as much knowledge as possible into it.

7. AI-Powered Email Filtering and Response

• Problem: Difficult to find important emails in a business inbox filled with spam and out-of-office replies.
• Solution: Use a Zapier agent to filter emails, ping the right people on the team, and auto-draft replies.
• Process:
  1. The agent filters out spam and out-of-office replies.
  2. It pings the right people on the team in Slack if there's anything important.
  3. It auto-drafts replies in the email inbox.
• Benefits:
  • Saves time and effort.
  • Ensures that important emails are not missed.
  • Automates email response.
• Key Argument: Improves email management and customer support.
• Technical Term: Zapier Agents (AI-powered automation tool).
• Real-World Application: The Rundown has landed two sponsors through this system.
• Prompt Example: Drop the provided prompt into Zapier's co-pilot to automatically generate the connections.

Conclusion

Rowan Chung leverages AI agents and tools across various business functions at The Rundown, including content creation, operations, and customer support. These AI-powered workflows enable a small team to operate with the efficiency and output of a much larger organization. The key takeaways are the importance of strategic AI implementation, the need for human oversight and refinement, and the potential for significant gains in productivity and scalability. The specific prompts and templates shared provide actionable insights for others to implement similar AI solutions in their own businesses.