Podcast Episode Summary

Podcast: How I AI
Episode Title: "Cursor is a much better product manager than I ever was": How this PM uses AI for PRDs, Jira tickets, and replying to coworkers
Host: Claire Vo
Guest: Dennis Yang (Principal Product Manager for Generative AI at Chime)
Date: October 27, 2025

Episode Overview

In this episode, Claire Vo interviews Dennis Yang, Principal Product Manager for Generative AI at Chime, about how he uses AI tools—primarily Cursor, an AI-powered IDE—not for coding, but as a daily workflow hub for product management. Dennis shares his practical tips and innovative workflows for creating PRDs (Product Requirement Documents), managing Jira tickets, replying to coworkers, and even prototyping AI products, all within Cursor. This episode is packed with real-world examples and actionable advice for PMs and anyone looking to boost productivity using modern AI tools.

Key Discussion Points & Insights

1. Why Use Cursor for Product Management? ([00:10]–[04:29])

• Cursor is usually seen as an AI tool for coding, but Dennis leverages it as a writing and workflow platform.
• Key Advantages:
  • Access to multiple AI models (e.g., Claude, GPT-5, Deepseek).
  • Integration with various productivity tools (Jira, Notion, Confluence, Figma, GitHub).
  • File system access for managing documentation and artifacts.
  • Customizable with user-defined rules (Cursor rules) for workflow optimization.
• Dennis Yang:
  “Not only does it have the AI…it has all the interfaces and interactions and connections into the tools that are critical for my daily product management.” ([03:36])

2. Optimal Setup & Usability Tips for Cursor ([04:29]–[09:05])

• Bigger screens are better to accommodate multiple panes: chat, document, files, and settings.
• Markdown is essential—using extensions and preview modes enhances readability.
• Dennis:
  “AIs…love Markdown, right?...Writing in, you know, looking at Markdown and then previewing Markdown is really important to see what it looks like.” ([07:41])

3. Workflow Automation: PRDs to Collaboration ([09:53]–[16:11])

• Write PRDs in Cursor using Markdown, then:
  • Use Confluence and Notion MCPs to publish for broad organizational input.
  • Control drafts via Git, but recognize the need for accessibility and comments from non-technical stakeholders.
• Discussion around merging code and documentation within the same repo for better context and source-of-truth management.
• Dennis:
  “PRDs are typically a snapshot in time…what if I put a cursor rule in that says…remember to update the PRD with the latest of how you’re thinking about this.” ([12:56])

4. Interoperability & SaaS Competition ([16:11]–[17:00])

• Open access and interoperability between tools is crucial; avoiding data silos enables better communication.
• The best tools now compete with simple, source-controlled Markdown.
• Dennis:
  “Any system, if it feels like it’s locking that content or data away, I’m not going to prefer to use that system…that’s my content and I want all my systems to be able to access it when it needs to.” ([16:11])

5. Demo: Using MCPs to Push, Update, and Reply ([17:21]–[24:09])

• Pushing PRDs: Simple conversational commands in Cursor update external systems.
• Receiving & Responding to Feedback: AI reads, prioritizes, drafts responses to comments in Confluence/Notion; Dennis reviews before they’re posted as “him.”
  • Memorable workflow: Write PRD → Publish → Collect comments → AI drafts responses → Human reviews → AI posts as Dennis.
• Dennis:
  “It organized the comments into high priority, medium priority, and then actually wrote suggested responses…Since the MCP is authenticated as Dennis as me, it appears to my other product managers at Chime that I am responding.” ([22:00])

6. End-to-End Project Management with AI ([24:54]–[32:47])

• Creating Jira Tickets Automagically:
  • AI reads PRD, generates Epics and Stories, fills out descriptions, acceptance criteria.
• Automated Status Reports:
  • AI-written, tailored by user rules and Jira data.
  • Frees up PMs from boilerplate busywork, lets them focus on higher-value activities.
• Dennis:
  “Cursor is a much better product manager than I ever was.” ([30:17])

7. Impact on Team Collaboration & Culture ([32:47]–[34:30])

• Asynchronous, structured updates improve communication in hybrid/remote settings.
• Fun side-tip: Use MCPs to leave “thank you” comments or emojis on Jira tickets, reinforcing positive culture.

8. Daily Routine & Personal AI Workflows ([35:14]–[39:10])

• Dennis asks ChatGPT for a “morning briefing” every day.
• Observes the strengths/weaknesses of LLMs in personal productivity, memory, and context over time.
• Dennis:
  “If you’re not using it every single day, you will not notice when things get better or worse.” ([39:10])

9. Prototyping New AI Products Without Code ([40:36]–[46:35])

• Dennis uses Cursor as a no-code prototyping sandbox for agentic workflows:
  • Writes PRD → Has AI generate TDD (Technical Design Document).
  • Outlines stepwise instructions for the AI to act as a “Super MVP Agent.”
  • Chain multiple models and tools with natural language instructions—no code necessary.
• Dennis:
  “It’s not even no code, it’s zero code. The words are the code.” ([45:17])

Notable Quotes & Memorable Moments

• Dennis on AI’s Value in Mundane Tasks:
  “Ruthless prioritization is what good product management is all about.” ([24:54])

• Claire on PM Cognitive Load:
  “You just do all this translation and then you get to like Jira tickets or support documentation, and you’re like, I can no longer do a good job here…I have reached the limit of my cognitive interest in this task.” ([28:05])

• Dennis on Source of Truth and Interop:
  “The most useful solutions will have interoperability as one of the key things.” ([15:49])

• Claire on AI Models’ Personality:
  “It says like, why is this model being so nice to me today?...Like, this is exactly what I need.” ([39:34])

• Dennis on Zero Code Prototyping:
  “It’s a really fun way to work because this isn’t even no code, it’s zero code...The words are the code.” ([45:17])

Timestamps for Key Segments

| Segment | Timestamp | |----------------------------------------------|-------------| | Why use Cursor for PM, not coding | 00:10–04:29 | | Cursor setup, UI, and Markdown | 04:29–09:05 | | PRD creation, collaboration, and Git | 09:53–16:11 | | Interoperability and tool competition | 16:11–17:00 | | Demo: MCPs, pushing updates, auto-replies | 17:21–24:09 | | Jira ticket and status automation | 24:54–32:47 | | Organizational effects, culture, async work | 32:47–34:30 | | Personal AI workflows (morning briefing) | 35:14–39:10 | | Prototyping agentic workflows, zero code | 40:36–46:35 | | Lightning round: getting started, best prompts| 46:50–48:55 |

Practical Tips & Workflows

• Cursor as a PM Hub: Treat it as your IDE for all product artifacts. Use chat and file panes generously.
• Publish-Reply Loop: Draft in Cursor → Publish to Confluence/Notion → Receive company feedback → Have AI suggest replies → Human reviews and sends—all within Cursor.
• Automate Toil: Use AI for Jira ticket decomposition, status report generation, and routine updates.
• Prototype AI Products Without Coding: Outline agent instructions in natural language and test flows using Cursor’s built-in models and tool integrations.
• Keep Humans in the Loop: Insert your review process where it adds value; let AI handle rote and repetitive work.
• Embrace Asynchrony: Reduce synchronous status requests by generating structured reports and encouraging comments in source-of-truth locations.
• Personal Productivity: Use daily AI rituals (e.g., morning briefings) to sharpen your intuition for LLM strengths/limitations.
• Kind, Clear Prompts: Maintain a polite, patient tone—even with AI. Restart threads if things go wrong.

Final Thoughts

The episode showcases how AI (and specifically Cursor) can elevate a PM’s daily workflow way beyond just writing or summarizing text. Dennis demonstrates you can automate, orchestrate, and prototype complex product management workflows—and even interpersonal communications—by combining best-in-class AI models, tooling, and a no/zero-code approach. For PMs (and anyone in knowledge work), this episode offers a blueprint for integrating AI deeply and practically into everyday routines.

Dennis Yang:
“Cursor is a much better product manager than I ever was.” ([30:17])

Where to Find Dennis Yang:

• Twitter/X: @Sindh (Dennis backwards)
• LinkedIn: Dennis Yang
• Upcoming: Speaking at FinTech NerdCon in November

For more: Visit howiaipod.com or listen on your favorite podcast app!