Podcast Summary: “I’m incapable of doing my job without AI”: How this Top PM Uses Claude + ChatGPT as His Second Brain

Podcast: How I AI
Host: Claire Vo
Guest: Amir Klein, Product Manager at Monday.com
Date: October 6, 2025
Episode Duration: ~38 minutes

Main Theme

This episode dives into how Amir Klein, a top AI-powered PM at Monday.com, uses Claude and ChatGPT as his “second brain”—organizing work context, automating competitive research, improving his communication, and preparing for interviews. It’s a practical masterclass in operationalizing AI to boost efficiency, enhance personal development, and maintain superior product management flow, complete with demos and actionable tips for listeners ready to copy Amir's workflows.

Key Discussion Points & Insights

1. Why Amir Relies on AI as a “Second Brain”

• Context switching is hard:
  Product Managers (PMs) constantly juggle meetings, projects, stakeholder asks, and shifting goalposts.

  “You have so many things to juggle that I discovered... you can silo like brains of conversations and threads in GPT and Claude that it made me super dependent on it in order to just do my work.” (Amir, 03:37)

• Solution:
  Amir creates "fully loaded brains"—dedicated, context-rich spaces in Claude and ChatGPT for each initiative. These hold files, conversations, and instructions so he doesn't have to keep everything in his head.

2. Anatomy of a “Second Brain” for Product Managers

• Folders & Files:
  Each project gets its own folder with all relevant files—kickoff decks, PRDs, competitor docs, customer conversations, etc.

  “I have a bunch of them and in it we have my files and my instructions... and you can see my endless amount of threads that I have with it, like non stop.” (Amir, 05:04)

• Input is Everything:
  Nearly any document—marketing pages, PDFs, slide decks—can be uploaded and used as reference context for the AI.

  “Literally everything is text. So…download as a PDF and you have... whatever it is... here as a reference.” (Amir, 15:44)

3. Automating & Analyzing Customer Feedback Using AI

Problem: PMs need real customer data to fight bias and build what users actually want, but manually searching communities is inefficient.

Amir’s Workflow:

• Source unbiased feedback:
  Amir wanted to understand what users think about “AI agents” for Monday.com—without being trapped by internal company narratives.
• Reddit Scraper with AI Assistance:
  • Used Claude to guide him (step by step, as if “I’m five years old”) to build a Reddit scraper that gathered 34,000+ rows of real user conversations. (08:35–11:03)
  • Compared mentions of Monday vs. competitors.
• Data Analysis with Claude:
  • Loaded the giant CSV back into Claude, prompting it to summarize themes, using tables for frequency and representative quotes.
  • Confirmed outputs by spot-checking and requiring source quotes to ensure accuracy.

    “Whenever I ask GPT or Claude to analyze something, I say reference, give me at least one or two quotes as to where this came up from.” (Amir, 13:40)

Integration into “Second Brain”:

• All analyses, raw files, and summaries are fed back into the project’s brain as reference for future product decisions and internal comms.

4. Leveraging Custom GPTs for Concise, Impactful Writing

Challenge:
Amir’s feedback was that his (otherwise good) writing was often too long for Slack and meetings.

Solution Workflow:

• Collect Writing Best Practices:
  • Amir loaded content from expert newsletters and books on concise writing into a custom GPT, making it his “writing coach.”

    “I took these newsletters, I command peed it, downloaded them and I put it into a custom GPT as... my guidelines to how to write.” (Amir, 18:28)

• Process:
  • Draft a Slack or document message.
  • Paste it into the custom GPT, which rewrites for clarity and conciseness, keeps his natural voice, and avoids generic “AI-speak.”

“I ask that question and usually does a pretty good job…and here we go.” (Amir, 17:03)
“People respond to my Slack messages more now.” (Amir, 21:42)

• Feedback Loop:
  Claire notes that many people get feedback on communication but lack a daily workflow for improvement. Amir’s method is an actionable answer.

5. Using GPT Voice Mode for Interview Prep & Public Speaking

Problem:
Traditional mock interviews (YouTube, mirrors, partners) don’t give realistic or interactive practice.

Amir’s Solution:

• GPT Voice Mode as a Mock Interviewer:
  • Created a custom GPT with best practices for product interviews.
  • Loads prep docs, position descriptions, and even his resume for super-contextualized practice.
  • Requests that the AI “not guide,” be candid, and provide actionable feedback at the end.
• Benefits:
  • Highly interactive; feels natural (“supernatural,” Amir says).
  • Tracks improvement over multiple sessions (the AI comments when his “user segments” improve).
• Tip:
  Even in freestyle mode, give the AI clear instructions on tone and what you want from the session.

“I forgot that I'm talking to AI. I feel like I have this… interview coach who's with me at all times…” (Amir, 24:34)

Adjacent Use Cases Discussed

• Public speaking and presentation prep.
• Oral reading and dictation feedback for children.

6. General Reflections & Pro Tips

• Everything is text:
  Use any and all reference material—“the list is endless.”
• Fact-checking habits:
  Always ask AI for sources/quotes; don’t trust blindly.
• Seamless context recall:
  “The ability to be at so many places simultaneously” is the game-changer for PMs. (Amir, 36:11)
• Iterating on Workflows:
  Amir is working on a tool to let others build these AI-powered research scrapers more easily.

Memorable Quotes & Moments

• On dependence on AI:

  “I'm incapable of doing my job without AI.” (Episode title & Claire Vo, 03:16)

• On efficiency and knowledge:

  “I can be asked on the spot from pretty much anyone in the company about something and I can give you an answer really quickly. I couldn't have done that pre AI.” (Amir, 36:11)

• On feedback and improvement:

  “Pre-GPT era and my dependency on AI…I would write out a message to myself, send it to my manager and say, hey, tell me how this can be improved. And he wouldn't respond. He came back a couple of days later—by then like I lost momentum…this is a game changer for me.” (Amir, 23:01)

• On voice mode for interviews:

  “I will be that lunatic in my living room talking to AI as if we're in the movie Her.” (Amir, 24:34)

Timestamps for Key Segments

• Setting up “Second Brains” for Product Work: 00:00–06:25
• Scraping and Analyzing Customer Feedback with AI: 07:34–15:44
• Using AI to Improve Writing / Custom GPT Demo: 18:05–21:42
• Preparing for Interviews with GPT Voice: 24:09–31:49
• Biggest Advantages for PMs Using AI: 36:11
• Personalizing AI Behavior & Fixing AI mistakes: 37:16–37:52

Actionable Takeaways for Listeners

• Start building folders and project brains in your preferred AI tool to hold context, files, and decisions for each initiative.
• Automate customer research by scraping community sites and running analysis in Claude or ChatGPT (request tables and direct quotes for accuracy).
• Create a custom GPT or prompt set for concise, impactful business writing—use expert templates and self-review with every message.
• Practice interviews, speeches, or presentations with GPT voice mode; feed in specific context and ask for structured feedback.
• Fact-check all AI outputs with original references.
• Keep building—Amir’s “everything is text” mindset can turn any input into actionable AI context.

Notable Resources Mentioned

• Claude by Anthropic & ChatGPT
• Lenny’s Newsletter by Lenny Rachitsky (writing tips)
• Exponent, Ben Erez (interview prep)
• [Amir’s upcoming open-source scraper—keep an eye out!]

Closing Thoughts

This episode is a treasure trove for anyone overloaded by context switching, seeking to level-up communication, or prepping for high-stakes interviews. Amir shows, step by step, how treating AI like a thought partner (and building good habits around references, context, and feedback) transforms PM workflow—making you faster, sharper, and always ready.

Find Amir on LinkedIn, and keep sharing creative AI workflows.
For resources and show notes, visit howiaipod.com.

End of summary.