How I AI - “Cursor is a much better product manager than I ever was”: How this PM uses AI for PRDs, Jira tickets, and replying to coworkers | Dennis Yang (Chime)

Episode Date: October 27, 2025

Dennis Yang is the Principal Product Manager for Generative AI at Chime, where he’s pioneered AI workflows that meaningfully increase productivity. While most people use Cursor as a coding tool, Dennis has turned it into a comprehensive product-management system that automates PRD creation, documentation management, ticket creation, status reporting, and even comment responses—without writing code. In this episode, he shares his end-to-end workflow and how non-technical professionals can leverage AI-powered IDEs.What you’ll learn:Why Cursor is the perfect hub for product management (even if you don’t code)How to use MCPs (Model Context Protocols) to push content between Cursor, Confluence, and NotionThe workflow for creating PRDs in Cursor and automatically responding to commentsHow to automate Jira ticket creation directly from your PRDsA system for generating comprehensive status reports without manual workHow to prototype AI products in minutes using Cursor as a “super MVP” environmentWhy source-controlled markdown files might replace traditional SaaS tools—Brought to you by:Zapier—The most connected AI orchestration platformBrex—The intelligent finance platform built for founders—Where to find Dennis Yang:Twitter/X: https://twitter.com/sinnedLinkedIn: https://www.linkedin.com/in/dennisyang/Chime: https://www.chime.com/—Where to find Claire Vo:ChatPRD: https://www.chatprd.ai/Website: https://clairevo.com/LinkedIn: https://www.linkedin.com/in/clairevo/X: https://x.com/clairevo—In this episode, we cover:(00:00) Introduction to Dennis Yang(03:00) Why Cursor is ideal for product management workflows(04:53) Setting up Cursor for non-coding use cases with markdown preview(09:35) Creating PRDs in Cursor and using source control for documentation(10:33) Using MCPs to publish content to Confluence and Notion(11:38) Bridging the gap between engineering and product(17:00) Reading and responding to document comments with AI assistance(21:37) Creating comprehensive Jira tickets directly from PRDs(25:51) Generating automated status reports from Jira data(30:23) Building a morning briefing system with ChatGPT(35:03) Generating personal morning briefings using ChatGPT(40:04) The “super MVP” approach to AI product development(46:37) Lightning round and final thoughts—Tools referenced:• Cursor: https://cursor.com/• Confluence: https://www.atlassian.com/software/confluence• Notion: https://www.notion.so/• Jira: https://www.atlassian.com/software/jira• ChatGPT: https://chat.openai.com/• Claude: https://claude.ai/• Git: https://git-scm.com/—Other references:• News API: https://newsapi.org/• Semrush: https://www.semrush.com/—Production and marketing by https://penname.co/. For inquiries about sponsoring the podcast, email [e-mail].

We've seen so many people use tools like cursor to write code. We actually haven't seen anybody yet just using cursor to write. That's what you're doing. The reason why cursor is my favorite UI for the AI is it has all the interfaces and interactions and connections into the tools that are critical for my daily product management. I think we're at this really interesting place where because these primitives are being built in the context of software engineering, you're getting these concepts of markdown, Git, commit, change tracking in these tools that used to be very software engineering centric? The most useful solutions will have interoperability as one of the key things. So any system that if it feels like it's locking that contents or data away, I'm not going

to prefer to use that system. I don't care why you have these modes, I'm sure there's a good reason, but that's my content and I want all my systems be able to access it when it needs to. It's really improving communication. I'm reducing the time I'm spending writing status and at the same time improving the status content that is being circulated both up to leadership and across the organization. Welcome back to HowI. AI. I'm Claire Vout, product leader and AI obsessive here on a mission to help you build better with these new tools. Today, we have a fun conversation with Dennis

Yang, principal product manager for generative AI at Chime. Now, this one makes me sweat a little bit because I thought I was the Alpha AI Powered PM and Dennis shows me his workflows which are way beyond anything I've seen before. He's going to show you how to use cursor to not only write your PRDs, but push them into Confluence or Notion, read comments, reply to comments, prototype AI tools, and more. This is an awesome one for anyone out there who's curious if you can make use of cursor without writing code, and I think you're going to learn a lot. Let's get to it. AI is supposed to make work easier, but I've been there. Weeks of setup, endless back and forth with engineering and yet another tool the team never really adopts. That's why I use Zapier's AI

orchestration platform. It connects with nearly 8,000 apps, so I can finally put AI to work without the drama, without the delays, and without pulling engineering in every time I want to automate something. With Zapier, you can roll out AI-powered workflows that do real work across your whole company in days, not weeks. I use Zapier every single day. It automatically responds to leads with enrich personalized data. It checks my calendar weekly and offers smarter ways to manage my time, and it even drafts emails for every new request that lands in my inbox. All of that running quietly in the background so I can focus on the work that matters. And Zapier's built for scale. With enterprise-grade security, compliance, and governance, it's trusted by teams at Dropbox,

Airbnb, Open Door, and thousands more. Go to try.zapier.com slash how IAI to learn more about how Zapier can bring the power of AI orchestration to your entire org. Dennis, thanks for joining How IAI. I am really excited about this episode because we've seen so many people use tools like cursor to write code. We actually haven't seen anybody yet just using cursor to write. That's what you're doing as a product manager and somebody who's thinking about strategy all the time.

So before we dive in, why cursor? why writing, you're not writing a lick of code in here for at least this use case that we're going to see. So how did you kind of get into this flow with this AI powered IDE? To me, cursor, the reason why cursor is my favorite UI for the AI is a few things. It has access to all of the models that I want to try. So in cursor, you can talk to Cloud, you can talk to GP25, you can do deepseek or whatever. That's the first thing. And then the second thing is it has access to a file system so it can write things down.

And then it can have access to cursor rules, which you, as you're using more and more, you can start to tell it how you want to do things. And then the final big thing is interop. It actually works with all of the different tools that I need to work with. So that's through MCP. It can talk to Jura and notion and confluence, you know, look at Figma, all these things. So basically, not only does it have the AI, the UI for the AI, it has all the interfaces and interactions and connections. into the tools that are critical for my daily product management.

You know, and the other thing that I would say is running on the desktop is just fast. You know, it's like pretty fast at doing those things as opposed to, you know, I think we're in this real moment where there's going to be a question of, are we going to start seeing more and more desktop apps for these AI power tools just because of the performance side of things compared to web? Web is very flexible. But I like cursor because it is, it's zippy. So this is your new hub for.

for getting work done. This is what my screen looks like. And the other thing I'm noticing is that I want bigger and bigger screens. I'm running out of space because I want my chat, right? I want the thing, the artifact that I'm working on. And then I want the file system. And then over here, I actually have the, this is my settings pane dock to the bottom because you want to make sure that it's still green.

So I need more screen real estate. And then. Yeah, we had Lee from Cursor on. recently and he walked us through the three pain kind of model of cursor for those who are listening and not watching you have a file system on the left where you're looking at what files you can work with or in context of cursor in the middle you have your artifacts whether those are code code artifacts or content artifacts on the right you tend to have your chat and then something I want to call out for people is you also have this bottom pane you can pull up and you

I was going to say, me as a developer, that pain for me is always the terminal. But this is a really nice little quality of life hack that I'm going to start to use your bottom pane in cursor is actually the tools cursor settings. So you can turn on and off MCPs, which I actually do all the time and confirm that they're working. Yeah, exactly.

So sometimes cursor would be like, I can't do that thing. And you look and that's like a little red dot. And you can turn it back on. Yeah. Well, you know what? Before we dive in now that I'm, curious and staring at your screen, would you mind sharing what MCPs just generally you have

turned on? And then we'll go through the few that, you know, you really use on a daily basis, or at least in this workflow. Sure, sure. So I have MCP. I have two that talk to Alassian suite. So one is our Jira Direct MCP. This is before Elassian launched their MCP official server, first-party server. This is another one that I can just run locally. Notion, of course, It talks in MyNotion, Figma, if you have Figma running, and then GitHub talks directly. And then these two MCPs are, I'm actually going to talk about them. I'll talk about a news API, which is an MCP I wrote to talk to news API.

It's just an API that searches news articles. And then SEM Rush is like an SEM, a search engine marketing MCP that I put together as well. Using cursor to write the MCP and then you actually can use cursor to install the MCP onto itself, which I love too. AI all the way down, as we say. Amazing. Okay, so you've loaded up cursor, you have it set up to write documents. Are there any other kind of cursor set up steps that you've done that you think are worth calling out for folks that are using it not to code? Yeah, I think one big thing is as you're sort of working more and more, not just cursor, but just LLM's A in general loves Markdown, right?

So, like, I'm starting to think in more and more in Markdown. So I'm actually looking at something like this all day and writing in, you know, looking at markdown. And then previewing Markdown is really important to see what it looks like because this is actually hard to Grock, right? So one of the good cursor settings here is, I think it's, yeah, preview, mark, markdown preview is a setting that you can actually find. inside inside.

So you do command comma, markdown preview. And in here, you basically want to automatically preview markdown. And the extension that I'm using here, if you go to settings extensions, is called markdown, oh, there is, markdown preview enhanced. So that's one I really like. And this one will actually automatically show in the preview pane when you click on it. you click on it, which I absolutely love. Because it used to be that you had to have, you had to, you had to know this secret code, which is like command shift V to open up the preview. But now you can

just click on a Markdown thing and it pops in automatically. Yeah, I just want to call this, call this out for folks who I know everybody, everybody loves Markdown now. It is the year of Markdown for sure. And Markdown can be rendered in a nice, in a nice way. You don't have to look at a bunch of hash headers and weirdly formatted table content. So either by installing an extension here or using that magic key combo there, you can get nicely formatted markdown right here in your cursor view. So, okay, we've done the setup. We've talked about extensions, cursors, your panes, your MCPs.

Let's actually get into how you use cursor as a hub for workflow, how you would achieve something as a product manager sort of end-to-end using cursor in these tools you've attached. I think it's been well-established that LOLM's AI is really good for writing PRD. So you should be doing your PRD creation in whatever flow that you want, be it chat GPT or Cloud or chat PRD, right? And for me, I actually do a lot of my writing. Now that I have this cursor set up, I do it directly in cursor. And it's working on, you know, in Markdown technically.

And I can preview it here. But here at Chime, we actually, you know, it's not a single player company. And I think that's one thing that I really want to be pushing on here is that the reason why you want to be interacting with all these other systems across, you know, not just your AI, is that there are other people usually in your company that you want to be working with. So we have a ritual here at Chime, which is when PRD is an early draft form, we'll share them out into our PRD draft channel and then gather tons of comments. So I work inside cursor. I'm making my PRD here. If it's in an early stage, I don't want some comments. I'll throw it into the PRD draft channel.

I do use Git to source control everything, but Git is not great for the whole company to go into and make comments. So I actually use the Confluence MCP to publish both into Notion, and into confluence. Because some people love using notion, some people live using confluence. So then we have, let's see here. So we have, here's this exact purity

pushed into confluence. I pushed it into this last night and threw it into our comments channel and already starting to get comments. Well, what I want to call out really quickly before we go into maybe how this, how this MCP actually works because I think people just like to see it is one you know we we talked about using an IDE as basically your text editor which is really interesting you're not writing directly in

confluence or Google Docs or anything you're writing in an IDE in Markdown with a nice preview the second thing you call out I'm curious your thoughts on is get and source control for non code assets or not explicitly code. Assets, I think we're at this really interesting place where because these primitives are being built in the context of software engineering, you're getting these concepts of marked down, Git commits, like change tracking in these tools that used to be very software engineering centric. Yes. And I'm curious, just like let's take a minute as a product people. Do you think we're going to bridge that gap? Do you think there's going to be like Git for, well, maybe I should build it, but like get for PMs or do you think we're still a little, like the abstraction is

too high for it to be useful and you need some of that UX polish that you're seeing in some of these more classic productivity tools? I have a lot of thoughts about this. So as we're sort of creating these artifacts that inform the product that we're building, traditionally today these artifacts sit outside of the code base. So this PRD is sitting inside. Comptus. So this PRD is sitting inside confluence right here. But now what I'm realizing is since I'm using cursor and using get to source control

to this artifact, it's actually now sitting inside this artifacts directory. And what I'm wondering is that if I think I want to start to see if the artifacts are actually sitting inside the repo into which the code is being developed,

that adjacency actually encourages the engineers and the AI coding assistant to continually have access to this. You don't have to give it an MCP to confluence. This is the source of truth. So you can read it.

And when things happen, so PRDs are typically a snapshot in time of this is what we thought that things should be at this moment. In development, we constantly learn and iterate and learn to iterate, but rarely do we go back to the PRD to edit it. So what if I put a cursor rule in that says, hey, if I'm working on this feature

and I'm learning things about the feature as I'm working on it, remember to update the PRD with the latest of how you're thinking about this. That's how the chat PRD repo works. So we have, amazing. We have a, we have a docs directory inside our sort of like repo for chat PRD. It has a combination of product documentation and engineering documentation listed.

There, even to-dos, like we've even pulled some of the like just task track into the repo because then, you know, everybody can work on it, AI or our human engineers. And so we've actually moved all of our, all of our documentation, PRD's, technical documentation and user-facing support docs into the repo because, you know, when I'm working on the product, I'm 90% of the time just working directly in the code. And the code needs the context of the documentation, the documentation needs the context the code. And so I think it'll just be really interesting to see where the source of truth content pieces lie. I mean, the other thing that's kind of interesting is because you're in

like a cursor or something that is an MCP client, you could just put the link to Confluence in there and then it could call the MCP and look it up on this other source of truth. So, you know, who knows where all this data will accrue. There will be battle of the enterprise software source of truth platforms. But I do think there's a lot of flexibility and optionality. And I do really believe, just like you said, code and documentation are going to start living in the same place more and more. Exactly. And that brings me to another thing that I'm really realizing is in this future, the most useful solutions will have interoperability as one of the key things. So any system that if it feels like it's locking that contents or data away, I'm not going to prefer to use that system, right?

Yeah. Like I don't care why you have these modes. I'm sure there's a good reason. But that's my content. And I want all my systems be able to access it when it needs to. Yeah. So I think that's going to be increasing more important. Well, and what's really interesting is, you know, you've been in product management for a long time and B2B.

And we used to always say MB2B, like your number one competition is like an Excel spreadsheet or a Google sheet. Completely. And now like your number one competition is like a source controlled markdown file. Like if you cannot outperform the source controlled markdown file as a as a as a as a as offering, then you're not adding enough value. So I think it's going to be really interesting. Okay. Apologies to to our audience. You got two product people in here.

I have a lot of thoughts about where SaaS is going. of AI products. Let's get back to the workflow. So you're making your PRD's in cursor. You use the MCP. And it would be just really helpful for people, you know, who haven't used an MCP before. Just to show how simple it is to do something like this. So let's say you wanted to push this PRD into notion or into confluence.

Can you show us how that works for people who don't ever seen it before? Right. So this thing is great. Actually, this I need to tell us. Look at that exclamation mark. You're so polite. No, no. So let's publish it into Confluence.

And I already pushed it, so I'm going to tell it into... Yeah. Don't overwrite the other one. We already pushed, make a copy, please. And I'm demoing this capability. All right, so we're demonstrating this capability. So this is basically all that's easy. So you always want to look to see that your MCP server is green.

It's planning. It's planning. It's moves out. And then basically just this is it. Like it's just running. It knows all of its tools. One thing, if you haven't seen how MCPs are set up, if you click on tools enabled,

you can mouse over each one of these and you can get an understanding for how each of these tools works or what they do. And for me, just when I show people, the MCP descriptions here, you start to form a mental model on what tools you've given your AI and what it could do. And how to ask for it. And so the other thing is not only how to ask for things from your MCPs, so really understanding the naming of the tools, but the other thing is to look across your tools. And I'll tell you all a little painful issue that I ran into is a lot of SaaS products or SaaS products have things called projects or files or docs.

And when I say update the project, it's like, do you want me to update the confluence space? Do you want me to update the chat PRD project? And so I think this ability to toggle on and off tools in the context of what you're working helps it just narrow in on what task you want to do. So a little lesson learned from Claire and named, poorly named, in specifically named MCPs. Exactly. The current MCP server seems to be having issues. is this what we get for doing it live yeah well and and here's here's the thing that you all you know here's the learning how i i i my couple things i think mcps are really opaque to people who yeah have not set up their first one to set one up you're usually given at least on a hosted mcp a URL you paste a couple

lines of config into cursor or whatever your client is it'll like boop you to log in to the app just like you would sort of ow off another app or put in an API token and then that she gave you access to it and then you do exactly what we're watching Dennis do right now. If you all want to know the state of this highly stable technology and platform, which is you toggle this button on and off over and over

again till it turns green. So we're going to give you the benefit of the doubt and say what really is going to happen is it's going to call these tools. One of these tools is going to say push a confluence document to a space and then that's going to show up in your confluence space as you proven in-house. This is the thread

of conversation. This is what it should look like. You say publish this PRD into confluence, and it affects it who writes it. So when the MCP works properly, and then in my PRD itself, I actually have a place where it can put its own published URL, right? Yep. So this is what it should look like once you've published it. It really is this easy once the underlying plumbing is working properly. Yeah, I do think that's worth calling out which is it's not just one way with these MCPs. So you again showed, hey,

I can take this content that I have in cursor. I can push it to confluence. But then you can actually do that round trip of context back into what you're working on in cursor and say, great, add the URL to that page in my doc and update everything. And so it's just a very nice new user interface. This used to have to be buttons and complex API mappings.

And now you have this. you know, language model that can figure out all these parameters for you based on very high level context, do the work for you, kind of self-heal on issues, and go back and interface with other tools as well. So I think it's a really nice flow. So other than kind of creating PRDs and pushing them into notion and or confluence and how, how product managery of you that they have to go both places? They're both great tools. Yeah. What else are you doing with these MCPs in terms of like project management, project workflow, collaboration? You know, we love to share our PRDs across the company to see what comments.

This is the thread where I'm now gathering, like I said, looks like people have been commenting on the PRD. Please go through the comment one by one and let's see how we'll respond. So what the MCP actually does is it reads the PRD. It sees the exact comments. So, I mean, I could use my own eyes and read the comments myself. and respond to them here in confluence, but that's no fun. So we'll have cursor read the comments, and then it actually does a breakdown. And this is, I found quite hilarious.

It organized the comments into high priority, medium priority, and then other different types of clarifications, and then actually wrote suggested responses, a lot of which sounded decent. So I started saying, like, yep, that one's okay. You can respond to that one. And since the MCP is authenticated as Dennis, as me, it appears to my other product managers at Chime that I am responding to this way. Okay.

Okay. So this is true behind the veil stuff. I have not seen this before. And this is spoken from somebody who has thought a lot about AI generated PRD. So just to clarify for people who are not watching, Dennis, Dennis writes. his PRDs in AI, he pushes them into notion or to confluence with AI. He then waits a little bit, reads the comments with AI, AI generates comments back, and you review them and you either approve or not, and then they get posted as you. Right. So here's... I love this. I love it. Yes,

that's exactly the flow. And I am, I'm in the loop, right? Like, one of the key things, building with AI is that you want to insert the human at the appropriate point where they make a lot. They're adding the most value, right? So gathering the comments, that's not value add. But reading them and understanding them and responding, like, I can provide perspective there. Yeah. This is really interesting because I do feel like people are really good at that push part of creating an asset and pushing it into another system. but I haven't seen somebody sort of close the loop on the next step and iterate forward.

I also think this shows that, you know, you're a busy person. And honestly, people's, you know, the comments need you to think about them and agree with a response or not, but you don't have to type with your human fingers every single response. and you can also prioritize what you want to focus on. And I think that's really, really valuable as well. Exactly. I mean, ruthless prioritization is what good product management is all about, right? I love it.

This episode is brought to you by Brex. If you're listening to this show, you already know AI is changing how we work in real, practical ways. Brex is bringing that same power to finance. Brex is the intelligent finance platform built for founders. With autonomous agents running in the background, your finance stack basically runs itself. Cards are issues, expenses are filed, and fraud is stopped in real time without you having to think about it. Add Brex's banking solution with a high-yield treasury account and you've got a system that helps you spend smarter, move faster, and scale with confidence. One in three startups in the U.S. already runs on Brex.

You can too at brex.com slash how I.A. All right. So what's next after you're gathering comments? Your PRD is great. All of your compatriots are loving the PRD that you've created and commented on. Typically, the next step in our workflow is to get this out into Jira. So read the PRD and create an epic. in the ATIA project.

So I created ATIA automation is our GRRA project that we're gonna be using. I'm gonna make sure this is green and then I'll watch it work now again. So what I really like about this is that when I'm doing this, right, it's reading the PRDs. I don't know what all of my other fellow product managers are like, but when Dennis creates tickets in Jira,

I could probably do a better job. But when Dennis's cursor creates the tickets, Dennis's cursor reads the PRD that I've spent a lot of time doing and then splits the effective tickets. The story tickets in particular are very, very well described. So I love doing the style of ticket creation. Because it really, it does what you wish you could do if you had infinite amounts of time. It's your words, it's your PRD, it's your stories,

and it's putting them where they should go, which is in the actual story tickets that then go to the engineers. And then they can, you know, you can always just say read the PRD, but they have to read the whole PRD to let's create story tickets. Let's get this ready and when it's ready.

That are all, let's create story tickets that are all associated with the parent ethic for this feature. And again, this is sort of a conversational flow. I've in my, in my personal cursor, for product management cursor roles, I'll say things like remember to always associate story tickets with their parent epic.

I did notice that when I was doing this at first, and I didn't say associated, then it would just create orphaned tickets without association. So this is the flow, right? It's handling a lot of the busy work, right, that I don't get a lot of value from doing. Well, and I want to go back to what you were saying, which is this is one of those tasks I say that is such a toil reducer for product managers

where you really hit cognitive fatigue on translating the same content for different audiences. And like PMs out there, we feel you. We know what you do. You take your customer notes and you turn them into a PRD and then your engineers don't want to read the PRD so you make a one pager and then the one pager has to turn into an email for your boss and the email for your boss needs to be three. bullet points for your CEO. And like, 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. And so you get lazy. And you like, kind of like do the epic name and the ticket names. And then you say link to PRD in the, in the description. And you push that cognitive load on another person in your team. And so I think one of these tasks that are kind of like administrative, low, you know, low incremental value is a really good thing to offload something to chat, or, you know, to Claude or AI with. Completely. A lot of this like housekeeping rig and roll, like these tools are so good at. And then what's, as you're saying this, like, what's another thing that product management

just constantly do, which is that same as X shape, we update status. So how might I use cursor to update status? So here's, here's epic. Let's pretend I'm one of the engineers and I pick up one of these tickets. I comment and look at the, and look at the ticket has. I know it's their sound noise. It has GERCEN. It has acceptance criteria.

I'm like, I built an agent or two that does this in a couple other platforms. than Jura. And it's just like, everybody's like, oh, I got a good ticket. I got definition of done. It's organized.

So this is a better job than I would do. I'll tell you that. I mean, this is, cursor is a much better product manager than I ever was. So, and now we're going to make it done. Right?

So we're going to do a few tickets in here. Again, once again, what is a status report other than using a tool and doing a job? So let's, once again, it's time. For status reporting, let's write a status report and describe everything that has happened since 924 for this epic. Give it this epic here. And what it will do is it will look at, it will write JQL and

essentially write a status report, right? And I've been doing this for about almost two months now, where every week, my weekly status report, I have a very long cursor rule now that I'm able to do this sort of repeatedly. Because one thing that you'll notice is, since I didn't give it what a status report was, it's going to just figure it out. This is like a zero context status report request.

And you likely have some ideas as to, what you want from your status report. So the recommendation here is you do it interactively first, and then at the end of this whole process, you would review the work and then give it feedback to do it better, and then you would save that into a cursor rule of this is my weekly status report, right? And what we learned when we started doing this on a weekly basis is that since the source of truth was in Jira,

like my engineers started commenting more in Jira about what was happening because they knew that someone was looking at it. at it. Right? And those words and that context was being added to the tickets. And even if you don't, even if you have a Jero ticket that simply goes from in progress to done and only has a ticket title and nothing in description, that's actually sufficient context to say that this thing went from this to done. Right. So it's really improving like communication. I'm reducing, I'm reducing the time I'm spending writing status and at the same time improving the status content that is being circulated both up to leadership and across the organization.

Well, and I've been, I've been to your office so I know you all have lovely and lots of nice time together. But I also want to call out for anybody who's working in the hybrid or remote environment. One of the biggest taxes on organizations is synchronous communication where like a PM is pinging an EM or an engineer being like, what's the update here? and in sort of like allowing those updates to go into a source of truth and then be queried in a really natural language way. Again, you don't have to change your behavior as a PM. You can still ask what's the update here, what the status, your life is better.

But the source of that data is more structured can be more asynchronous, I think allows people to like do less context switching, less synchronous, you know, communication, all those things that just give us more time to create. So I think that's a really interesting secondary effect of what you're showing here. Definitely. And even if you're doing an in-person or over Zoom kind of status update, again, we have all these tools to record, transcribe, document, and put all of this content back into where it should go, like hang it off of the geo-ticket, and then it's all organized for AI. And now you've given me like maybe this is like a chaotic good idea,

but I was like, oh, you know, as a PM, you really got to get people to like you. Like, this is one of these things. Pro tip. PM's got to be likable. And I'm like, oh, man, you can use the cursor or that, not the curse to MCP at last you and MCP to like put nice comments on Gira tickets that are done. Right?

Like, thank you. Thank you. Thank you. I do a lot of thank you emojis. Yeah. Right? It's the fabric of culture that we have to establish across. Yeah.

It's the fabric of culture and also just powered by a commercial large language model. This is where we're at. I love this. I mean, this is a really interesting example because, again, it doesn't matter with the content of what you're working on is I think everybody can take away. I can write an asset. I can push it to the right sources. I can query the comments. I can query the feedback. I can translate that for my team on whatever format they need. And I can get aggregate insights and updates. This is awesome. Awesome flow has given me a lot of ideas that I am going to take. So we have one more quick, kind of like a little bit more. personal workflow that you were going to show us. So let's bop over to that and see how you start your morning. Everything's a morning briefing with you, I guess. Yeah, exactly. This is how I, so how do I start my day? One kind of fun thing that I like to start everyone with when I'm talking to them about chat GPT is you go into chat chit and all you say is write me a morning briefing based on what you know about me. And chat chatt chit has added fantastic memory. to its system, such that its morning briefing is actually pretty decent.

So how I begin my morning is with the chat GBT, every morning, it essentially compiles for me what it thinks I'm going to be interested in. And I've been doing this for a long time, and every morning at around 5 a.m., it automatically creates this morning briefing for me based on what it knows about me. So you can see here, we had an earthquake, the national stories. And what I did notice is after a while, it's starting to actually starting to lose the plot.

So I need to, instead of informing it by just memory, I'm thinking now I need to give it more context in specifics of what I want. So this to me is actually, it's informative in understanding that even open AI is not perfect at figuring out what memory context is relevant for a very, very small, you know, request like morning briefing. But one of the key things that I really tell everyone in terms of how to, how do we learn how to use AI? You have to use it in this way. And you can kind of understand and get a gut feel for how it does a bad job sometimes. It really helps you understand how to make better AI product. And is this a custom GPT? Do you just literally write morning briefing? Is it just a long

standing chat. What is it? This is a chat GPT project into which I've placed, I've given it no help. Oh, okay. I have not told it anything around like what it should be doing or even given the files. Because I'm actually trying to

see how it does with only the previous threads that we've been doing. Yeah. Interesting. And then you just say morning briefing and it goes. Yep. It used to do a fantastic job. I would say about a month ago. It did a great job every morning.

It's feeling like either it's losing the plot and this project is getting too big or some sort of model changes. I'm not really sure. Yeah. Oh, no, those are bullet points. So that's GPT5 right there. Yeah. But I do like how, you know, they're putting these little screenshots in, which is nice.

But this is like, no, there's no AI news anymore. And like, where did my AI news go? Maybe the takeaway for folks here is, you know, if you're trying to get into AI, I think for two reasons. One, if you're trying to get into AI for personal productivity reasons, find a daily use case. It's like really simple, consistent, adds little value in your life just because it'll add a little value in life. I think the second order effect of that, though, is somebody who's thinking about AI is it's a really sort of natural, repetitive way to learn about the strength, weaknesses, evolution, skill sets of these core models on which almost every product, you know, for the next couple of years is going to be built. And so you can start to form your own intuition as a product person of, okay, like, why is memory failing here?

Why is context failing here? Where do instructions really help? Is my two-word prompt, what I have to call you out on? That suggests as morning briefing. Like, is that sufficient? I'm coming in and I'm typing this in every morning. I should probably test the like schedule repeated task thing. And so again, I just think get in there and find something that every day you're going to find useful so that,

you're getting used to these tools and you're really understanding what their capabilities are. And then if you become somebody who's going to build these tools, you're a little bit ahead. You have more language about what makes a good user experience with AI. Exactly. If you're not using it every single day, you will not notice when things get better or worse, right? You know what? This is very true because I open cursor today and I don't know if this is an effect of I really did not sleep well last night. But cursor is like so chipper today.

It's like real friendly. I think also because I'm like Claude Sonnet 4 today, which I rarely use. Oh, really? It's like such a cheery little, cheery little guy. Oh, yeah. I like like a grumpy, like a GPT5 middle of the road, like factual staff engineer or just like a clinically depressed Gemini 2.5. So I'm used to those. And so I was like, why is this model being so nice to me today?

Like this is exactly what I need. And then I was like, oh, it's sweet little claw. And unless you talk to all of these different models, you won't have a mental model of what their personalities are like. All right. So you showed us how you can just create this and kind of like prototype this morning briefing in chat GPT. But I know you. You're an AI PM. You want to build this thing. And I love the way you prototype your actual AI products and agenetic products in cursor. So let's whip back to cursor. And I want to I want to see. how you would actually go about as a PM building that kind of product and prototyping it and testing it in cursor in a really lightweight way. That's a great. I love this question because I always tell people you should be prototyping all of your AI product

ideas inside chat GPT first, right? You should just try it. That effectively was my morning briefing prototype was chat GPT. And now I use cursor itself to continue to prototype in what I call typically a super M3 MVP, super minimal. And the reason why it's super minimal is because I'm using cursor itself, which is AI, to prototype the AI product that I'm about to build. Does that make sense? Yeah, yeah, yeah. Yeah, because cursor has access to all these models. So you don't have to set up too much to at least get the baseline sort of act. So how do we do this? So we have our PRD

that we just wrote. Here's the problem. Here's some solutions. I have it write a TDD for itself, here's my TDD. I say you're going to be a super MVP. You, Cursor, are going to help me prototype and understand how to do this. And just for folks that don't live in three-letter PM word, TDD is a technical design document. Yes. So Cursor itself will write an approach, a technical design document to prototype this product that I'm trying to build using AI, which is a morning briefing system. So what the Super MVP instructions now are is it's going to create instructions, and this is what we're looking at now, for Cursor to do the task that I want it to do. In this case, you can see here, step one, load the configuration, step two, we have profiles about what news to look for.

Step three, use the MCP news search tool to search for news, and then process the content, summarize it, and create a report. So this is effectively what some are calling prompt engineering, but instructions for the AI to do the job. Yep. That looks all great. Now I can just say at super MVP agent instructions run today's briefing. Yeah. Go. Make sure my MCP is still green while this is running. Looks good. And there it goes, right? So basically it's creating this report that I just defined in our PRD, allowing me to prototype it. It's making, I love their new to-do function. I love it.

I love it. If you notice here, this is, it's running date. Amusingly, sometimes cursor doesn't know what date or time it is and it will argue with you about what date it is. So I told it, make sure to run the terminal date command. Yep. So you can check what date it is. And this effectively is doing the whole job right now of going to the news at MCP, gathering news that I'm interested in, it's going to read the output, and then generate the report. And right now, this, I'm using CloudForce on it.

The amazing thing about this is I can use these same exact instructions and have GPT or Gemini or, you know, any other model run it and see how each model differs. Okay. So this may be super nerdy, but this is blowing my mind because if I, if I am somebody who builds AI products and I am somebody who thinks like prototype in a chat GPT or some like consumer general product and then pull it in. And the silly thing that I would do that you're opening my eyes to is I would go write code. I would like go write type script or Python or whatever and I would call in these like libraries, these SDKs. And I would structure my function. and I would like do all this work. But what you're showing as a hack is cursor actually has built in tool calling, model switching, like agentic workflows. And so bypass all that rote setup for yourself in this prototype phase. Just provide like the natural language description of the tools and functions or MCPs and let cursor do the dirty work of. calling the right models, chaining the thought, all that kind of stuff. Then when you feel like

your instructions and your general structure, right, great, somebody can write some code. But this is just like such a perfect intermediary step and such a hack for this like all purpose platform for AI that that cursor is going to become. That I think is just really clever. So good job, Dennis. Thanks. This is, it's a really fun way to work because like this isn't even no code. It's like zero code. Yeah. There's no, I guess it is no code. It's really no code.

It's not writing any code. Like, no code systems are typically write code, but this is, there's no code. This is the, the word, the words are the code. Well, okay, and this is not as, at as catchy as vibe coding, but we're like, we're, like, vibe agenting. We're, like, writing a pretty complex, like, agent setup. And instead of, like, buttons and prototypes and, you know, forms and flows, we're just getting kind of like an AI-powered experience.

And now we're seeing here, your morning briefing. And this is it. And I mean, basically, like, we're, we're so close to just shipping this thing, right? Like, I can, this is the content. This is today's, this is today's news brief for me. Yeah. And this is a place where then a PM can come in with opinionated defaults on, I think this should be Claude Sonnet.

I think, you know, make sure you're calling it a daytime tool. Like, here's the MCP that I set up for myself to pull news. like if not this one, another one. And so it allows you to get that next step on implementation as opposed to offloading that to the engineering team. Yep. Who might not be as opinionated on the user experience side. That's right.

Love it. Very good. Bravo, my friend. Okay. Dennis, this is so fun. I have so many ideas. AI all the way down. I have a couple lightning around questions for you and then we'll get you out of here.

Okay. One, do your call. know that you're replying to their comments with AI? Or is this the spoiler alert? No, I don't know. Hello, fellow PMs. 100% I think they all know. I think the one single benefit of us being returned office in person is that before that,

I don't think people believe that I was actually a real human. So they all know that I'm fully AI enabled. Perfect. I love it. Okay, my second use case is, or my second question for you is, how would you recommend a PM get started with cursor? Are you just like literally opening up a cursor and saying, hey, this is going to be a directory for product documents and artifacts? There's going to be no code here, set it up. Like, what's your zero to one quick start? Yeah, my zero one quick start is basically you open it up and you make a brand new directory just called like blanker in that case. And then you just start talking to it. As soon as you do that, you'll have a chat pane.

Right. So and then you can start learning how this actually works. Amazing. Okay. And then my last question, my favorite one, you were so polite to your AI. So I know this will be good answer. but when, for example, you're toggling on and off this MCP or it's calling the wrong tool or it's overriding your epic tasks with the wrong thing. What is your prompting technique? What's your go-to?

I think most people that know me know me to be a very kind, nice person. I don't, I'm not a yeller at, I don't yell at the AI. I'd say a lot of pleas. I say, as you noticed, I'd use a lot of exclamation points. and sometimes I just kind of start the thread all of them. Just like, all right, let's just start new, right? Completely time travel back in time and be like, let's start over. Maybe you need a break. That response has just like soothed my inner child. So I love to hear it, very calm, very positive, very kind, and just walk away when you need to.

That's a wise response. Okay, Dennis, this has been so great. Where can we find you and how can we be helpful? You can find me. I'm online typically as Sind, Dennis Backwards. Twitter, X, and then LinkedIn, you can find me there. I'm just Dennis Yang. And I'll be speaking at the FinTech NerdCon conference in November. Trying to gather as many AI-enabled people there. So come and join me there. Amazing. Well, thanks for being here. Really appreciate you walking us through this.

Thanks for having me. This is fun. Thanks so much for watching. If you, enjoyed this show, please like and subscribe here on YouTube or even better, leave us a comment with your thoughts. You can also find this podcast on Apple Podcasts, Spotify, or your favorite podcast app. Please consider leaving us a rating and review, which will help others find the show. You can see all our episodes and learn more about the show at how IAIIPod.com. See you next time.