Interview with Claire Vo, CPO at LaunchDarkly, about ChatPRD, an AI-powered product management tool she built solo over a Thanksgiving break as a side project. Now used by over 10,000 product managers with six-figure revenue. Topics include how she built ChatPRD with AI, a live demo of the tool and its document editor, how it fits into her workflow, AI's impact on the future of product management (PMs as "proto-managers" building prototypes), how junior PMs can upskill with AI, and using AI as a parent. # Channel Presentation Observations — "She Built an AI Product Manager Bringing in Six Figures—As A Side Hustle - Ep. 24 with Claire Vo" ## Show Identity & Format - **Show name (self-described):** The outro explicitly names the show as **"AI and I"**: *"you absolutely positively have to smash that like button and subscribe to AI and I."* - **Episode numbering:** Title includes **"Ep. 24"**, indicating an episodic interview series. - **Host:** The outro identifies the host as **Dan Shipper**: *"a journey into the future with Dan Shipper as the captain of the spaceship."* The host also self-identifies his affiliation: *"I actually have… at Everywhere… we build little products. We have a couple incubations going right now"* and *"we at Every… have a media business… we've already spun out an AI writing app called Lex."* ## Intended Audience - **Product managers, product leaders, and builders using AI.** The conversation is framed around product management workflows (PRDs, roadmaps, milestones), e.g., defining a PRD "for people who don't know what it is." - **Founders and indie/solo builders interested in AI-native businesses.** Extended discussion of bootstrapping, distribution, and whether venture funding fits small AI products: *"I don't know if the current model of venture funding is the right model for something like this."* - **AI-curious professionals and parents.** Segments on personal automation (email, school summaries) and kid-focused AI projects (Greek mythology cards, runawaypancakes.com) broaden toward a general AI-in-life audience. ## Claimed Scope - **Interview show about how people use AI in work and life.** Demonstrated by the structure: guest demo → work use cases → personal/family use cases → future outlook. - **Live product demos are a demonstrated format.** The guest shares her screen to demo ChatPRD, Linear, and Zapier automations. - **Audience call-to-action:** The closing is a direct subscriber pitch — *"Hit like, smash subscribe"* — with the show positioned as delivering *"pure unadulterated knowledge bombs about ChatGPT."* ## Creator Self-Description vs. Demonstrated Presentation | Type | Evidence | |---|---| | **Explicit self-description** | Outro calls the show "the epitome of awesomeness," a "journey into the future" about ChatGPT/AI knowledge. | | **Demonstrated pattern** | Long-form, conversational guest interviews with live screen-share demos; host references his own ventures (Everywhere/Every, Lex) as context, showing the channel sits within a media/incubation brand. | | **Guest positioning** | Guests are framed as practitioners at the intersection of leadership and AI side projects ("this really interesting intersection… building this side hustle that is built with AI and uses AI"). | ## Observable Tone & Style - **Enthusiastic, informal, exploratory:** Host reacts in real time (*"Wait, can I pause you right here… how is this built?"*) and thinks aloud (*"I'm actually thinking through real time"*). - **Practical/show-me orientation:** Repeated requests to see tools ("Can I see the Linear?", "show us some of your historical chats"). - **Promotional outro (possibly comedic/satirical in register):** The closing monologue is effusive to the point of parody (*"Dan, I'm absolutely, hopelessly, in love with you"*), suggesting an intentionally over-the-top signature sign-off. # Delivery Observation: Ep. 24 with Claire Vo ## Overall Format - **Long-form conversational podcast** with a live screen-share demo component. The guest walks through her product on-screen while the host interjects with questions. - The episode closes with an unusual AI-generated outro: > "Oh my gosh, folks, you absolutely positively have to smash that like button and subscribe to AI and I." and > "Dan, I'm absolutely, hopelessly in love with you." --- ## Tone ### Guest (Claire Vo) - **Casual, self-deprecating humor** recurs: > "I'm just a write your own sequel kind of girl." and > "That's how I use chat GPT is I write horrendous sequel and I say, 'Make this sensible or stop giving me an error.'" - **Playful demystification of technical work**: > "Document editors are hard to make. No you know here's what like is so fascinating. No they're not. Like this is a solved problem." - **Enthusiasm markers**: frequent phrases like "really fun," "really cool," "truly a lifesaver." - Frames AI as a companion: > "it's a little buddy that I can have that can just like round out the the edges of my thinking." ### Host (Dan Shipper) - **Curious, interrupting interviewer**: > "Wait, can I pause you right here for for 1 second? Like this is really beautiful and like well-built." - Frequent affirmations: "That's amazing," "That's really cool," "That's the best." - Summarizes and reflects back: > "so basically like in the process of of doing these yourself, you realize ChatGPT is really useful." --- ## Vocabulary & Recurring Phrases ### Guest - **Product jargon used repeatedly**: "PRD," "spec," "scaffold," "milestones and sequencing," "user stories," "functional requirements." - **Emphasis through "truly" and "really"**: "I did this truly on a Sunday," "this has been like just truly a lifesaver for me." - Repeated product habit: > "I'm going to be really lazy and say Teams um, plan features for ChatPRD." - Repeated shout-outs to tools: > "All the startups are getting shout outs. Clerk is so good." and "I'll give I'll give a start out a startup a shoutout. I'm using TipTap editor." ### Host - Heavy use of **"like" as filler** throughout: "like you're you're kind of in this really interesting intersection of like..." - Repeated launch phrase: "That's really cool" / "That's amazing" after nearly every demo moment. --- ## Teaching / Demo Habits (Guest) - **Live walkthrough pattern**: states intent → runs the tool → narrates output → critiques it. Example: - Intent: "Help me write a PRD Teams plan features for ChatPRD." - Critique: > "Okay, this is an okay start, but I actually feel like the user stories are really lame." - Follow-up prompt: > "expand out the user stories, focus on a few personas, billing lead, team lead, admin." - **Explicitly teaches a quality criterion** — specificity: > "I think it's um I think it's specificity" ... "instead of saying we need roles and permissions it says we need these three roles and permissions." - **Numbers as evidence**: "10,000 users," "50,000 chats," "285 chat PRD chats," "I've done a release every weekend since I launched." - **Narrative anecdotes with concrete timestamps**: > "I raised my hand in the meeting, I remember at 10:00 a.m. And then by 2:00 I had this like full five-page spec." - **Structured "three things" framing**: > "I I personally use it for writing PRDs for ideas I already have or brainstorming features on the road map. And then as a product leader, I use this to help me improve an existing PRD..." --- ## Communication Patterns - **Guest often self-corrects mid-sentence**: > "like all good workers I'm kind of work I You know, I maybe not like all good workers." - **Host and guest both use trailing open-ended sentences** ending in "like..." inviting the other to continue. - **Guest anticipates objections and answers them preemptively**: > "And then people are going to say, 'Oh Claire, but then you're ruining jobs for junior folks.' That is actually not true..." - **Uncertainty / hedging appears in live-demo moments**: > "It's slow. Sometimes docs are incomplete." and > "It's apparently not running right now." --- ## Stable Patterns vs. Isolated Moments ### Stable (repeated across the episode) - Guest's **critique-then-iterate** demo loop (used at least three times: user stories, timeline, technical considerations). - Guest's **self-deprecating framing of ease** ("really lazy," "not that hard," "solved problem"). - Host's **affirmation + follow-up question** rhythm. ### Isolated / Low-Confidence - The **AI-generated outro** appears once; no evidence elsewhere of this being a recurring device (no other episodes observed). - The **"bucket war" anecdote** and **Santa voicemail story** are one-off personal stories — amusing but not representative of a pattern beyond a general habit of family-related AI anecdotes (which do recur ~3 times: Greek mythology cards, podcast hub, recipes site). > **Note on uncertainty:** filler-word frequency ("like," "um") is qualitatively evident throughout, but exact counts were not tallied; claims above are based on repeated direct examples, not exhaustive statistics. # Observation Document: Claire Vo Interview (Ep. 24) ## Speakers - **Claire Vo** — Chief Product Officer of LaunchDarkly; founder of ChatPRD (identified in transcript by host) - **Dan Shipper** — Host (identified in outro) ## 1. Origin Story of ChatPRD Claire Vo's stated sequence: - **Initial use of ChatGPT**: As an executive leading product/engineering organizations of "many hundreds of people," she had output work beyond meetings, so AI assistance was "very welcome." ChatGPT "did not scare me. It felt like magic." - **Prompt development**: Over months she refined prompts with ChatGPT (GPT-4 era) to write product specs rapidly. Example: raised her hand at 10 a.m. for a complex audit tool spec; by 2 p.m. had a "full five-page spec" that stunned her team. - **GPT store launch**: Published her prompt as a ChatPRD GPT; shared with her team, "they all loved it." Joked: "it is just a prompt, but she's my prompt." - **Domain and app**: Bought chatprd.ai for ~$60, put up a newsletter signup. With no monetization path in the GPT store — "I'm the kind of product person that needs to make money" — built the app herself over Thanksgiving week, launched November 28, initially at ~$2/month. **Metrics claimed by Vo**: ~10,000 app users, 50,000–100,000 chats on the GPT, six figures of ARR at six months, 285 personal ChatPRD chats in 6 months. ## 2. Key Claims About AI-Assisted Building - **The "aha" wasn't generation but recall**: AI flagged functional requirements she would have forgotten ("You need filters"), acting as "a little buddy... that can round out the edges of my thinking." - **Building is cheap now**: "It is so cheap and easy to build high-quality web apps right now... it's a database on a website." Host expressed skepticism ("Document editors are hard to make"); Vo countered: "This is a solved problem" — she built Notion-style collaborative editing (via TipTap) "on a Sunday... while my kids were watching Aladdin." - **Stack**: Next.js + Tailwind; coded largely via ChatGPT from ChatPRD-generated PRDs; uses Stripe, Clerk.dev, Metabase (SQL), Linear, Slack. - **Solo founder quality thesis**: Building solo "does increase quality and velocity" by "short-circuiting the loop between product engineering and design" — one person holds vision fidelity rather than deciding by committee. Shipper agreed via analogy to multimodal models. - **Speed multiplier**: Vo claimed building her pre-AI SaaS startup took 5–9 months for a first version; with AI, "if you gave me a coffee the size of my face... I would have it done by Sunday afternoon." Qualification: she *could* have built pre-AI ("I'm a serviceable but not excellent software engineer"); AI provides an on-demand version of mentors (Yaelin, Jeremy, Dave) she previously learned from in person. ## 3. Demo: Product Workflow - Live demo generating a Teams-feature PRD for ChatPRD; iterative refinement — criticized initial user stories as "very, very high level" and pushed for **specificity** ("instead of saying we need roles and permissions it says we need these three roles and permissions... I can hand this to an engineer"). - Chat-to-document mode: documents regenerate (not yet inline-editable) — an acknowledged limitation ("we're getting there"). - She described herself as "pretty lazy" in prompting because she prompted ChatPRD to ask clarifying questions when context is insufficient. - **Actual top use case** (her claim): "the number one use case is engineers and founders that don't have a product person." - Cadence: "I've done a release every weekend since I launched" — framed as commitment to velocity and proof that a serious executive can still build. ## 4. Future of Product Management (Vo's Positions) - **PM-to-builder ratio shift**: Cited a user claim that one PM can now lead 20 engineers vs. typical 1:7–1:10; that user skipped hiring an APM in favor of a designer. - **Junior roles not ruined** (her rebuttal to anticipated objection): junior PMs ramp up faster via on-demand coaching, which executives lack capacity to provide personally. - **Strategic thinking is disruptable**: Vo stated LLMs are good at synthesis and applying pre-existing frameworks, so strategic thinking is "very likely to be disrupted" — and she deliberately uses AI for strategy "because that's the part of my job that's most likely to be disrupted, so I better get really good at it." - **What remains human**: inspiration, alignment, motivating teams toward a mission; long-range vision-setting (possibly from design/engineering leaders); the "proto manager" concept (friend Waylon's idea: prototype managers who build prototypes instead of documents). - Shipper agreed LLMs handle "well-trodden paths" but humans retain "force of will or the vision to be like this can work, and I want to do it." ## 5. Personal-Life Automations - **Recruiter email automation**: Gmail filter → Zapier → OpenAI endpoint → drafts polite declination with future job preferences → Slack recommendation. (She notes it did *not* get her current job — a recruiter's personal text did.) - **School email summarizer**: daily 3–4 p.m. summary filtered to her kids, flagging action items ("picture day is tomorrow"). Called it "truly a lifesaver"; would teach others "as a public good" rather than productize. Includes self-acknowledged privacy tradeoff: "I try to be protective of information, but... I got to get these summaries." - **Kids' projects**: Greek-mythology Pokémon-style cards, a podcast quiz hub, runawaypancakes.com (kid-followable step-by-step recipes with Midjourney images). - **Shipper's contribution**: trained an ElevenLabs Santa voice for behavioral voicemails; his kids now demand custom stories — "in one ways it's very magical in other ways like spoiled for choice." ## 6. Business Model & Funding Positions - **Goal**: "I want to see if we can get it to a million dollars" — kid-centric milestones (wine → school tuition → college 529). - **Venture skepticism**: She *could* raise (friends would write checks; host offered one), but questions "does this kind of product need that kind of check?" She's probing "how far you can take this with me and Alyssa, who works part-time." Qualification: "never say never... if 10,000 users turns to 100,000 users in the next 6 months... we might have a different conversation." - **Venture-outcome logic**: "$100 million is not a venture outcome" — taking VC commits you to that path; she knows "that game" from prior fundraising. Project is profitable and self-fundable. - LaunchDarkly is aware and supportive; IP carve-out in place ("we're all good corporate citizens"). - Shipper concurred: venture model was built when software was riskier and outcomes bigger; his org (Every) runs a ~$1M/year business on ~6 people with AI leverage and is weighing "the shape of that money." - **Voice-mode strategy example**: drafted her company strategy via voice ChatGPT during a 22-minute Bay Bridge commute. ## Preserved Tensions & Qualifications - Host skepticism ("Document editors are hard to make") vs. Vo's "solved problem" rebuttal — left as stated disagreement. - Vo's pre-AI capability is affirmed *and* the AI speedup is claimed as "thousands of times faster" — both positions held simultaneously. - VC path is rejected as "not obvious" while explicitly left open under growth conditions. # Speaker Attribution & Provenance Record **Video ID:** 01a0f15b-0c00-7df7-99d3-e5052ca47b4f **Title:** She Built an AI Product Manager Bringing in Six Figures—As A Side Hustle - Ep. 24 with Claire Vo ## Speakers & Roles - **Dan Shipper** — **Host / creator** (implied by outro: *"a journey into the future with Dan Shipper as the captain of the spaceship"* and channel name **AI and I**). Conducts the interview, asks questions, shares anecdotes (negotiation deal, kids' custom stories, Santa voicemail, Lex/Every incubations). - **Claire Vo** — **Guest**. Introduced as **Chief Product Officer of LaunchDarkly** and **founder of ChatPRD**. Primary source for most content: origin story, live product demo, personal automation workflows, family AI projects. - **Unnamed third parties referenced (not speaking):** - **Alyssa** — works part-time on ChatPRD team. - **CJ at Artisan Talent** — recruiter who connected Claire to her current job. - **Waylon** — friend; source of the "proto-manager" concept. - **Yaelin, Jeremy, Dave** — engineers Claire previously learned from. - Claire's husband and children (Henry and Theo) — referenced only. ## Promotional Segments (Flagged) - **Outro call-to-action:** closing monologue urging viewers to "smash that like button and subscribe to AI and I," ending with *"Dan, I'm absolutely, hopelessly in love with you"* — clearly a stylized promotional outro. - **Guest self-promotion (in-content):** Claire promotes **ChatPRD.ai** and her social handles (X: Clarvo, TikTok: Chief Product Officer, ChatPRD on Twitter/LinkedIn/YouTube) in the closing segment. - **Product shoutouts by speakers:** TipTap editor, Clerk.dev, Stripe, Notion, Linear, Metabase, Zapier, ElevenLabs, Midjourney, Greeking Out podcast — endorsements embedded in conversation, not paid segments as far as the transcript shows. ## Referenced Sources & Third-Party Claims - **ChatPRD usage stats** (Claire's own claim, unverified): ~10,000 app users; 50,000–100,000 GPT store chats; six figures ARR at six months; launched November 28; release every weekend. - **User testimonial quoted by Claire** (third-party, secondhand): PM claiming ability to lead 20 engineers using ChatPRD; "saving me 10 hours of work in 15 minutes." - **"Bucket war" historical anecdote** — Claire offers to provide a link; host confirms *"we'll put it in the show notes"* (link **not present** in transcript). - Mentions of prior employer ("my previous company"), a pre-AI SaaS startup, and Everywhere/Every's AI writing app **Lex**. ## Provenance / Availability Notes - **No curated metadata supplied** with this source; all facts derived from transcript alone. - **Comment availability:** not indicated in the source material. - **Timestamps:** not present in transcript; segment order follows transcript sequence only. - **Uncertainty:** speaker turns are not labeled in the transcript; attribution of host vs. guest lines is inferred from conversational context and the outro naming Dan Shipper. ChatPRD is a on-demand CPO that helps you with your product work. Help me write a PRD Teams plan features for ChatPRD. So then I just click go. And so it's going to go, "Great, here it is." Gives me a problem statement, business goals, user personas, user experience, success metrics. I think this looks great. I want to build this all in 3 weeks. So document's been saved. So if I click document save, now I have it in a document format. We have some of these AI features. You did this all yourself? Like this is complicated. Document editors are hard to make. Here's what is so fascinating. No, they're not. Like I did this truly on a Sunday, Notion-style editing in a day by myself while my kids were watching Aladdin. Clare, welcome to the show. Hi, I'm so excited to be here. I'm really excited to have you. So for people who don't know, you are the chief product officer of LaunchDarkly, a feature management and experimentation platform. And you are also the founder of ChatPRD, which is an on-demand chief product officer that writes and improves your PRDs. Um and we met at a sort of like angel investing retreat and I was like doing bits about AI and we just got into this like really great conversation. I feel like you're you're kind of in this really interesting intersection of like you're you're leading um a business as uh as a chief product officer and then you're also like building this ChatPRD side hustle that is built with AI and uses AI. And so you I think you just have so many like ideas going on in your brain about like what the future of this looks like. Um and so I'm just super excited to have you on. Well, I'm excited to be here. It's something that I think about a lot and as I was preparing for this conversation, it's something that I took a moment, reflected, I use a lot. So, I'm excited to just share a little bit about what I think and what I'm building. That's awesome. Um so, why don't you like give us a little bit of like a background on ChatGPT? Cuz that's the thing I'm like most interested in. It It seems so cool. Yeah, so um you know, like all good workers I'm kind of work I You know, I maybe not like all good workers. I'm trying to work myself out of work. That's That's the ideal way to do things. And so, um when ChatGPT and some of these tools came out, I was all over it. Um it did not scare me. It felt like magic. And I spent a lot of time as an executive leader, I lead product and engineering organizations, many hundreds of people, pretty large responsibility. Um but I still have believe it or not, more than meetings and like hiring great people to do. I actually have to output work. And so, um with the a calendar like mine and the demands of a job like mine, anything that could be a help it is is very welcome. So, I started using ChatGPT to help me um basically write product strategies and product specs. We were a fairly scra- I would call it a scrappy team um at my previous company. And occasionally, I would uh ICPM or more technical products. And so, something would come up, it would be pretty complex, it would span product and engineering, and I would raise my hand and say, "Hey, I'll I'll write the spec for that. I think I have a sense of what we need to do." And there's this very specific example of us building a pretty complex and custom um audit data audit tool. And I said I raised my hand in the meeting, I remember at 10:00 a.m. And then by 2:00 I had this like full five-page spec. And my team was like, "What just happened? How did you do that? Cuz you've been in meetings all day. Like, what was that?" And it was because I had over the course of of months, you know, sort of like prompted ChatGPT into a a place where I could really work with it in a pretty rapid fashion to get high quality outputs. That wasn't just going to happen with just using um kind of plain ChatGPT, like GPT-4 at the time, I think. So, when the GPT uh store came out, I thought, "Okay, I'm just going to like dropped my You know my joke is it is but it is just a prompt, but she's my prompt." And so, dropped my prompt into the GPT store, um got the like great name, Chat PRD, which I think is just It's good stuff. People love it. And shared it with my team. And I was like, "You all can use this if you want to know how I do it." And they all loved it. And so, I was just kind of joking around to my husband. I said, "We should just buy like we should just buy the domain on it." Of course, it's got to be .ai. So, you know, I spent my 60 bucks or whatever cuz there there's a premium on these AI domains. Bought Chat Chat Chat PRD.ai and just put up a like a newsletter sign-up form and a link to the GPT. And so many people started using it. And um you know, still with the GPT store, I think it's in early early days. So, even though it's getting a lot of use, there was no moneti- monetizing things. And I'm sorry, I'm the kind of product person that needs to make money off the things that they build. So, over Thanksgiving last year, um between, you know, like hosting the kids and and doing stuff when they were napping, I like, you know, dusted off VS Code. And I was like, "Yeah, I think we can I think we can build this." And so, I built it over the course of Thanksgiving week with my kids home while they napped. We can talk about all the ways I used AI to make that really, really easy. Um and launched it, I think, on November 28th. So, she's she's 6 months old. Um and just put like, I don't know, it was like $2 a month. I was like, "I don't know. Like, who knows what somebody's going to pay here?" And people started buying. And then, I started to add a little bit more features and raising the price, and people started buying. and then now I have, I think, 10,000 users on the app app um that I host. Over 50,000 chats have been done on the GPT. It's somewhere between 50 and 100,000, because I think that's where they like bump you to the next the next tier. Um and I started to build out a lot more than just chat features on the product. So, it's been really fun to build, and I do I think I have 285 chat PRD chats in the last 6 months. Um so, we use it a lot. That's wild. There's so many things I want to dig into there, but I want to just like sort of summarize the like process that led you here, cuz I think it's a sort of generalizable thing. I actually have a I actually have something similar in my own life that I've done with this, where it's like it seems like you you used it yourself, and you built up a very complicated prompt um that reflected like your worldview and your skills and and all the ways that and your process for like making PRDs. Actually, before I keep going, can you define a PRD for people who don't know what it is? it's I I now say uh Chat PRD is a on-demand chief product officer that helps you with your product work, cuz we've really expanded past PRDs, but a PRD is a product requirements document. It's definitely It's uh generally a doc or a spec that defines the problem space that you're going after when you're building a new product, kind of what users want, if you have one user or multiple users of the product, what are their incentives, what do they want, what are their goals, what are their non-goals, um and then kind of details out some of the features and capabilities you would need to make this meet the needs of their goals, and then a lot of um teams and companies use it also to outline things like tracking plan, security risks, technical considerations, marketing campaigns. So, it's sort of the source of truth for when you're building something new as a product, whether it's a very small feature or a brand new company, sort of um the written record of of that's going to be. Okay, that's great. So, so basically like in the process of of doing these yourself, you realize ChatGPT is really useful. You You create these prompts that help you build good PRDs. Um and then you realize like oh, this is sort of like complicated for me to like prompt manually all the time. And then you like start just like making a GPT, and then you share it with your team, your team loves it, and then you're you're sort of off to the races like okay, I'll I'll have it on the store, maybe I'll build my own version of it. Um but I think that that's that's the sort of like repeatable process that I think a lot of people are finding is off on your own explorations, make something awesome, make some awesome prompts, and then you can sort of like productize that. Yeah, and I think the the key that we're not talking about is like how do you distribute it? Because unless you are maybe featured in the GPT store, which can get you distribution, you really like you own your own distribution for even the GPT store or an app. And so I was lucky enough to have um a good number of people who follow me for product content, so it was like right in in the wheelhouse of who might find it useful to use the product. And then it creates that virtuous cycle of word of mouth distribution. But I do think distribution still is a challenge for anybody kind of going down this path. Yeah, that makes total sense. Um and so just to like just to back up to sort of that the beginning of the story, like tell us about like what you were using it for initially. Like what were these sort of initial prompts that you were you were working on? Like what were the tasks that you like or the mini tasks within making these documents that was actually helpful for you? Yeah, so basically we have um a PRD template. So, we have like every time you're going to build something, here's what we just generally expect you to outline and write. And then as people write those, then we have um at my old company we call it PRDR, product requirements document review meeting, where we actually like debate uh both asynchronously in comments and then live sort of the the the thinking in this document and then it becomes a source of truth for what to build. So, we have this template and PMs, including myself, would, you know, once a week sit down and be like, "Ah, I got to pick up that next project." and sit down and start with a blank, almost blank, slate, a template and go, "Okay, I got to write this whole thing out." and they end up being somewhere between um a mini PRD would be just like a one page up to like seven, 10 pages just depending on the complexity of the product. And a lot of times you're doing things like it it needs to have role-based access and like these very non-specific but functional requirements need to be outlined and if you forget them, then engineering goes off the rails and you get a product where you forget to build a whole bunch of table stakes features. And so, what I would do is I would paste it and I was like, "This is our template. This is like the type of company I'm working at." and then just word vomit, here's the product that I think we need to build and why, and it would scaffold out then all those sections combining the context of the template, what I've said about the business, kind of like high-level what I've said about the product I want to build. And the thing that really was the aha moment for me was not that it could do that. Obviously it could do that. Um it was that it thought of functional requirements I had I would have truly forgotten it. And I think I'm pretty good. I've been doing product for 20 years. Like I am pretty good. And when I was building this audit feature, I was focused so much on the data we needed to ingest and how people needed to audit. And then um when I was generating this PRD via AI, it said like, "You need filters." And I was like, "Of course I need filters. Like obviously I need filters." And somebody probably in a meeting would have said, "Clare, you need filters." But I could have just But it put it in the doc. And it was just one of these moments where I was like, "Oh, it's a little buddy that I can have that can just like round out the the edges of my thinking, make me a lot more efficient, and remind me of things even that I need to include where my fallible human brain fails me. I love that. I think that's I think that's really common, too. It's like um some of the stuff that you learn from it, it's not stuff that you wouldn't already know. It's just like you might not have thought of it. It like finds the simple things that you're missing and that is sometimes so valuable, especially if you're working alone. Like you're you have a team and stuff like that, but like, you know, building chat chat PRD where you're it's like I think it's it's you and maybe one other person part-time. Like it's sometimes hard to uh it's hard without a buddy to remember that that stuff and and chat chat chat GPT is that. Um I'm really curious like to get into like how you're using it. Like do you want to show us some of your historical chat PRD chats? share my screen. Okay, so this is this is chat PRD and some of the the things that I think are really um unique compared to the GPT store is how one not only I've customized it, but you can actually customize it to yourself. So, you can tell um Wait, can I pause you right here for for 1 second? Like this is really beautiful and like well-built. Like how is this how is this built? Uh I did it. I mean, this isn't magic. Uh if anybody's building stuff right now, I'm going to be like it's a Next.js app, like mostly Tailwind. I have good taste. I don't know. It's pretty I I think one of my other learnings here is it is so cheap and easy to build high-quality web apps right now and as somebody who's built web apps many times in their lives and run very large engineering teams that build web apps, um it's actually not that hard to build something pretty phenomenal, you know, it's a database on a website. So, thank you. I appreciate that. I I try to I try to put love in it. We can talk about how actually I think um building solo does increase quality and velocity in many ways because you're short-circuiting the loop between product engineering and design in a way where like I think in a traditional team where you have multiple people playing that role, you lose fidelity of kind of vision. And so one of my one of my interesting ideas is which has something like Chat PRD like plus a co-pilot or like a dev in, do you actually collapse and focus the ability of one person to build something and does the quality actually go up because you're not negotiating these you know you're not uh deciding by committee, you're deciding kind of like one person. So thank you. everything integrated. It's the same thing with like the multi-modal models. It's like a lot of the stuff gets better when you integrate the like the the image recognition and the video into the intelligence portion and it's like yeah, like integrating development, design and product thinking into like one person is going to like is you're you're faster cycle. So you're going to it's going to be it's going to be higher quality. I think that's really cool. And did you use Chat GPT to code this or did you code it by hand? I used Chat GPT to code it. I mean look, I can write some code but it's much easier to say I mean I take a a PRD from Chat PRD and like hey, this is this is the component I need to build. Can you just scaffold it out for me? And then I can play with the edges on design or functionality and um that kind of stuff. So of course. Okay, that's great. Okay, so let's go back to before I interrupted you. So like you're you're in settings. Yeah, so you can customize your profile. So one you can tell it a lot about yourself, who you are and what you do. It remembers that so it has memory. So unlike you know well, I know Chat GPT has memory now but a lot of times previously when I was building this I'd be like tell me about your role in product, you know, tell me a little bit more about your company. And so we're just stabilizing that information. And then as I said, most companies have a a template. I have a pretty good one that I think I use as a default obviously because it's mine but you can also add your own template of PRDs. Let's say you're like Um, you have a TLDR which I always do, short summary of the product doc. And then you have functional requirements. Are super important and, you know, put functional requirements in XYZ format. Whatever it is. Um, you can actually use a custom template here. This is something that's just really useful for people as they're using this in the business context to get consistency and again not have to manage that that prompting. Um, but what I use it a lot for our two things. I use it to, well, maybe use it for three things. I I personally use it for writing PRDs for ideas I already have or brainstorming features on the road map. And then as a product leader, I use this to help me improve an existing PRD to coach my team on how they can improve their PRDs. Um, so those are sort of three things. But let's just talk about help me write a PRD. So, one of the things that I've been working on is a Teams feature for Chat PRD. So, right now Chat PRD is sort of like in single player mode where individual PMs can use it, but we've gotten a lot of feedback that people want a um, Teams version of Chat PRD. So, I'm going to say help me write a PRD. I'm going to be really lazy and say Teams um, plan features for Chat PRD. Um, if I can go. Okay. So, then I just click go. And um, it's going to say usually do you normally do that level of detail or you normally like writing more stuff when you're actually doing this? pretty lazy. I usually do this. Because I got a prompted Chat PRD to like ask for stuff. Um, if it doesn't have enough information to do a good job. So, you can see it's like, okay, well, that's great, but what do you want? And so I say I've gotten a lot of feedback that Teams want to use Chat PRD PRD together. They have a few priorities, shared billing, shared templates, and stored um company contacts knowledge, Mhm. sharing chats and doc uh documents with each other. Uh if you have other ideas, please share them. I think we can grow to 100 teams accounts by the end of the quarter. Okay. It's like very very high level. Mhm. Um and so it's going to go great. Here it is. And so gives me a problem statement, business goals, user personas, team lead, um user experience, like how you might onboard team members, how shared billing, all this stuff works. Um a narrative which uh this is a little thing that I've injected here, which is I feel like PMs are very good at pitching their story. Um so I've added that in, success metrics, technical considerations, milestones and sequencing, etc. Okay, this is an okay start, but I actually feel like the user stories are really lame. So, I'm going to say expand out the user stories, focus on a few personas, billing lead, team lead, admin, individ- individual team users. But before you before you hit enter, I'm just sort of curious like um what do you like as a product leader, like when you're reading these, like what is standing out to you as like the user stories are sort of like lame? Um it's just very very high level. Um so they're just yeah, obviously, but then \{comma\} how? Mhm. Um and so I just think yes, I want to share chats and documents with my team so we can collaborate efficiently. Let's get much more specific. Okay. Um actually let's get into a little more detailed of a CV. A lot more detailed um outline features and sub-features that might be useful to all these personas. Um that's great. Great persona. I think I'm asking cuz like for me sometimes it's hard like seeing something like that from from uh from ChatGPT or honestly from people that work for me. Like I can kind of tell that it's kind of high-level but it's hard for me to like have that like I my eyes can just sort of glaze over and I can skip it and be like yeah like that kind of like Yep. that kind of makes sense. Like and I'm just sort of I want to know what's in your brain to be like okay no no no we got to we got to push in there. Yeah, yeah. I think it's um I think it's specificity and I think this really shows itself up in the in the user experience. Now we've gotten something a lot more detailed on onboarding team members where it's like okay we're going to invite them via email. Mhm. There's going to be a welcome guide. There's a secure link. Um there's roles and permissions. Here are actually the roles and permissions that you need to build. So this to me feels much more specific. I'm like oh okay I can hand this to an engineer now. Yeah. And instead of saying we need roles and permissions it says we need these three roles and permissions. Right. You can like see it. Yeah. You can see it. Yeah. And so here's where I think chat PRD the app gets really interesting. Um I think this looks great. I want to build this all in 3 weeks. Today is June 5th. Update the timeline. So then the the only other thing is I think is the the milestones. And so then they're going to it's going to regenerate this document, include all the updated um user stories which we have So now we have these more detailed user stories which I'm very happy about. And then it should generate um um more milestones in a in a more detailed way. And I'm I tend to be very aggressive, so I give it like I want to do this in 2 weeks or 3 weeks or whatever. Most teams tend to know how quickly they want to do things, but have a hard time like breaking down those milestones. And so I find I use milestoning and sequencing pretty frequently here as a as a structure or scaffold for how we might do milestones. So here like it's like initial development building features 10 to 13, 14 to 17, beta testing and launch. So it's got this nice like, okay, can I hold myself accountable to that? Can I build that into Jira. And then this is where we go beyond sort of what you could do in a GPT store, which is we have this idea of save as a document. Look at this nice little solo. It's slow. Sometimes docs are incomplete. I've apparently pushed I've talked to Open AI about this like their actual team and I've pushed the edges of some of their function calling. So we're going to see. We're going to wait for this to load. Um and while it loads I'm wondering if you have any any questions. I always have questions. Let's go forever. That's very cool. I was one of the things I was going to ask you is like um the decision to to have this be a chat feature instead of like like a document that you were progressively filling out like a document editor type interface. Um how did you think about that choice? Well, it's actually both. So if we can get this loader to to you know, if we can get the robots to spit back our document. I actually think multimodal is the way Okay, that wasn't too bad. Look at that. that was fine. Yeah. So document's been saved. So if I click document save, now I have it in a document format. I can actually toggle back and forth between chat mode mode. This is like the magic. And then then what I can do the document that it creates. So like if you go back into the chat mode and you're like, I want to I want to add something more to the user stories as well go and re like edit the document or does it regenerate the whole thing? It has to regenerate it right now cuz the way things work but we're getting there and we do we the here's the the proof that we're getting there. We have um some of these like built-in AI kind of features here. So if I'm like oh no we're hardcore we need this like very business. You did this all yourself? This like this is complicated this is complicated. Document editors are hard to make. No you know here's what like is so fascinating. No they're not. Like this is a solved problem. This is a solved I'll give I'll give a start out a startup a shoutout. I'm using TipTap editor a lot of a couple of AI companies using it. Like I did this truly on a Sunday. I built live it's real-time collaboration if you want it it's like ready to go. Notion style editing in a day by myself while my kids were watching Aladdin. Like I love it. There's just no I just tell people like there is no excuse not be able to build something. Um so yeah so we have this this document a badass like it's just it's like it's undeniable. No I just I have interesting hobbies but yeah we have this we have this document node node now where then what I do okay so that took me like five minutes to get all of this which we we love. And then I come in here and I'm going to be like uh I'm going to go down actually to um oh we don't have it in here. Let's add it in. So uh I don't have technical considerations so I would go back and be like I don't have technical considerations. And so it said hey making changes to the doc you can ask for more feedback. So the chat definitely has context of any changes I make to doc. I'm going to say I forgot to add we use Stripe billing and subscriptions. Can you add a section at accounts for the technical and implementation considerations for um billing. Also, I use clerk.dev. All the startups are getting shout outs. Clerk is so good. We use Clerk for some internal incubation stuff and like it's it's just you get so much stuff for free in the user management stuff that like would normally take you like even with AI it'll take you like a day or two to like integrate Google sign-in or whatever and uh and Clerk is just yeah, it's really good. Shout out Clerk. Yeah, so now I can tell it okay, now we're getting into like how we might actually build this and so it's going to take into account okay, what it knows about the Stripe billing capabilities, what it knows about Clerk um and actually give me some technical considerations that I think are really helpful when handing it off to an engineer. Spoiler alert, like I'm the engineer. Handing it off to yourself. Yeah, that's how that's how I might that's how I might structure the data or that's how I might do it. So here you go. Integration with Stripe, manage subscription plans for team, implement web hook listener, like all this stuff. Yeah, this is what I need to do. Um here's a question for you that like I'm actually thinking through real time cuz like we we at Everywhere like we build little products. We have a couple incubations going right now. Like are you using this um yourself for your for your for chat PRD? Like at what scale do you think like this level of planning is helpful um and uh and if you're using it for for your sort of like to help yourself track stuff like what you know, what does it do for you that um that you think it it it adds? So what I think is helpful, so I think it's very helpful at very small scale, single me, to very large scale, multi big team cuz I've seen it work at both sides. At the very small scale, this is what's helpful is there a lot of benefits to um working solo, a lot. But like if I were to this, I would like start in one corner and I'd be like, okay, let's like integrate organizations. And then I'd be like, what next? And then I'd have to like sit there and think about the um architecture and the sequencing and how I might do it. I would forget I'd be like, I never want to build a billing page. Like that's terrible. I don't want to do that right now. And if I don't want to do it right now, I'll never do it. So, what this gives me is like then I'll just go pop this and we're going to build an integration with like Jira, Linear, etc. I'm going to pop this. It's just going to go create like a project plan for me and at least I have my to-do list of things to do anchored on a document. Um so, as you can say see here, it updated it. It resaved the document and go back to it um and it should have technical considerations dropped in. It's like fun when you do a live demo. And then it works, yeah. Actually works. So, now I have all this and then what I would do is I would actually like share it um and I would copy the share link and then what I typically do is uh send it in Slack to uh Alyssa who works on our team and I'm like, what do you think about this? And she opens it up and gives me some opinions. When we launch actually the teams feature, we'll add commenting, we'll add like actual inner team sharing and stuff like that. That's really cool. Yeah, that's what I use it for. I love it. Um I I I'm thinking that we should be using this internally like cuz we don't we we use Notion which I also like Notion, but like but I think we could put these docs honestly just in Notion and we don't really have a we don't really have a set product management like flow cuz it's mostly just been me like building stuff myself until recently and now it's like we have a couple of engineers and and a couple other people involved and so it's it's getting to that scale and I'm one of my failures as like a product person is like I get too excited about whatever's new and then I and then I forget like the plan and I think having a plan having a plan would be really helpful for us. So, the really surprised so, built this like by product people for product people. That's how I built it. And actually the number one use case is engineers and founders that don't have a product person. There you go. And so like I have a bunch of engineers that are like I work with this cuckoo founder who just ideas. Not me. And they just built it, right? And so they like they take the idea and they dump it in here and they're like, "Okay, well like now I have a plan. Now I can build I can build this." And I've used this for everything from Let me see if I can find a good one. Um Oh, brainstorming features for road map. Mhm. So, uh I've This was uh actually how we built our summer road map is I said, "Here are my three big themes." People a lot of job seekers are using it. A lot of integration with productivity tools. So, we're talking about Google Docs, Notion, Linear. And storing more context for user research. So, these are things that people are wanting for us. And then it built out a bunch of ideas. And I said, "No, let's build at least five ideas with three pillars, business goal, etc." It built it out and then I saved that as a document. So, now I have like a road map. That's really cool. Basically, it's stuff to build. That's amazing. I I I don't know whether you've prepped to to show this. So, like if if for whatever reason you don't have it, it's totally fine. But like where is your like like dashboard hub for like chat PRD like internally? Like what are you using? And can you show it to us? Like how are you organizing all the information? Yeah, so what we use um is Metabase. So, I'm using I'm like just a write your own sequel kind of girl. I guess that's You know what? That's how I use chat GPT is I write horrendous sequel and I say, "Make this sensible or stop giving me an error." So, I use um I'll just talk about my stack. How about that? Um I'll talk about my stack. So, Oh, I meant I meant like just you know internally like all of your docs. Like your your where are you saving these docs and like and like planning things and and keeping things? Oh, yeah. So, we use Chat We use Chat Your PD. It's basically the source of all our product documents. We use uh Linear for um sort of like road map and feature tracking. Um and then we use Slack. Okay. That's really cool. Yeah. Um I I like Can I see the Linear or like is that is that is that cool? like it's not thrilling. You're you're welcome to see it. Uh sure, can you see my Linear? Um what is it? Linear. Uh Linear app. I'm like app.linear linear app. Okay, yeah, you can see our Linear. Um please, this is like fun stuff. Okay. So, teams accounts, account settings, core chat stuff. Fixed doc mode, this is like can I work around Open AI? Um we have bugs in general, improving onboarding and activation, integrations, marketing site. Uh this is their onboarding stuff. And then we have our summer road map, which I just shared in our Slack channel, so it's no surprise. I'm working right now on like doc mode being really great. We're doing some marketing site stuff. I'm doing some like back end billing stuff. Here's one of the things about being a solo founder on a thing you didn't think was going to be a real project. You get real lazy about like tracking billing. Sorry. And then I was like turned around and I was like, "We have thousands of users. I need to get better on like making sure subscription data and stuff is tracked in our database." Um we did Oh, actually did a really interesting onboarding if I can show it to you. Um and then we're doing a bunch of integrations in our backlog. So, this is our summer road map. We'll probably do a couple more things, but we track it here and then Alyssa and I I I truly I do a release I've done a release every weekend since I launched Chat Your PD. Every single weekend. It's like my commitment to myself and to my community and to my users. Like every weekend I release something. Um and part of that is one I just think product velocity really matters. And two, I'm trying to prove like you can do this I have a very serious I have like look at this suit. Like I have a serious job and I still think you can build like something pretty incredible and the cost of building is really pretty low. Um And so it's part of it is this like exploration of what it means to run an AI native business. And part of that is like I think building is cheap and you should prove you could do it Yeah. How different is it do you think like in in your view like could you have done this pre-AI? Um and and and what's the difference between you with AI versus you without AI in terms of your ability to like to build stuff like on the side? Um so I could I have done it pre-AI? Yes. Like I built I did a startup pre-AI. I was the you know only engineer on the team for nine months of the company. Like yeah, I could build a web app pre-AI. Here's what it is. I am a serviceable but not excellent software engineer and but I'm I'm a quick study. I'm very quick study. It's one of the things I'm good at. And the way I got to where I'm at from a technical perspective is I sat across from somebody and I was like, how do I do this? How do I do this? How do I do this better? How do I do this better? How did you do that? And now I just have that on demand. Like the you know, these people if they listen to this will know like Yashin or not Yashin, uh Yaelin and Jeremy and Dave. Yaelin and Jeremy and Dave, all at different parts of their career, sat across from me. And now I'm a monster. And so now I have like a combination of them you know, digitally sitting across from me at any moment. And so if I want to learn something new, um I can learn it very quickly and then I can put it into implementation without having to bring that like other person into the loop. Yeah, no, I that makes total sense and and yeah, I I think I'm I'm more asking like on the like on the side, like you have a demanding job, you have a family, like uh when you're talking about building things on the side, like what um uh would you have had the time to do this or and and if so, like how much faster does this let you go? No way. It's like thousands of times faster. I mean, I I I I said this yesterday at an event. I did a startup before. It was a SaaS app. It's a website on top of a database. It was like a good website on top of a good database, but it was a website on top of a database. And it took us probably five to nine months to kind of like build the serviceable first version of that. And I think if you gave me a coffee the size of my face and said on Saturday like go, I think I would have it done by Sunday afternoon. Like I cuz cuz you can stand up off, you can stand up a database, you can stand up building, you can do all these things, and then if I hit a wall, I go ask my like on-demand engineering buddy to write code for me and it gets written. So, I just I think like that is the multiples of efficiency you get. And if we believe you can get multiples of efficiency on engineering output, then I absolutely believe you can get multiples of efficiency on product output. And this is where like some of our great testimonials are, where people are saying like you're saving me 10 hours of work in 15 minutes. Like that's like great. Then use those 10 hours of work to do something really awesome for the company. Um so, I think it is a tremendous amount of leverage. That is that is really crazy. What What does that What does that say to you about like the future of product as a job and like where where does that go? What happens? Um so I I think there's a couple things that are going to happen. Um maybe one is that I think the ratio of PMs to sort of like builder roles is going to shift pretty dramatically. One of the Chapt users said to me, I can now basically run \{quote\} or um lead a team of 20 engineers, where typically like the ratios you're seeing in a lot of companies are like one to seven, one to like 10, maybe. So now you're going like one to 20. That's very different. And in fact they said, I'm going I can do one to 20. I was going to hire an APM to help me, sort of more junior PM. Now use Chapt here and I'm going to hire a designer instead, cuz I think that's the thing that's going to accelerate our our our our our building. Um so I think that's one thing. And then people are going to say, "Oh Claire, but then you're ruining jobs for junior folks." That is actually not true, because then what we're seeing on the other side is junior PMs are able to ramp up and be much higher impact, especially in a remote context, because again you have this on-demand coach that can help you up-level what you're doing, which is very hard. If you ever look at my calendar, like sure I would love to coach every PM one on the team, and I just do not have capacity to do so. And so I do think you're going to see like more junior folks have higher impact more quickly. I'm really excited about that. And then the third thing, I it's so funny. Um you know, there's a bunch of kind of leaders at my level, CPO, VPs of product, etc., that are like, "How is AI going to change things for them?" But like it's never like changed things for me. And I'm on the very strong opinion that one of the things very likely to be disrupted is strategic thinking as a differentiated value for a human like a human. I think that these LLMs are really good at synthesis, applying like pre-existing frameworks like business schools exist, they've published their information like pre-existing frameworks to business problems to drive specific outcomes. And so I use Chat GPT and Chat Purity a lot for strategic work. And I think that's partially because that's the part of my job that's most likely to be disrupted, so I better get really good at it. Um using these tools as opposed to sort of being on my on my back foot. I love that. I mean I I I love the sort of attitude is like don't panic, just use it. If it's going to be good at it, like just use it. Like I I can give up the the strategic thinking and and and and direct this thing to do it. What what do you feel like is most likely to be left for you? Like what are you what do you see the like PMs of the future being good at? Um yeah. I do think like inspiration and alignment and motivation of humans towards a goal is still just like it's it's very hard for for something like AI to do. Um getting teams excited about a mission, getting them close to the human impact of their work, which is like how it impacts customers. Um that's one piece that I think this sort of like influential and inspirational leadership you're still just going to have to have. At the PM level, I do think that the ability to build things is going to become much more important. And my friend Waylon has this idea of like the proto manager, which is instead of being a product manager, you're a prototype manager and you're going to be expected to build prototypes instead of building documents. And I think that's like really interesting thing to play with. So I you know, I think there's going to be like building is still going to happen on teams for a little bit um until these kind of tools get a lot more um consistent in their output. I think inspiration and leadership and honestly just convincing hordes of humans to do big things and be excited about it. Hopefully, I can still be good at that. And then I do think like these things are trained on past data. And so, what is I think there's going to be some product leaders, and I think they might actually come from the design side or maybe in the engineering side to be able to see very far into the future and sort of set a place to go. Now, somebody else can map how you get to that place, but really having that point of view of like what's the place you want to go. I do think there's some opportunity there for people to contribute. Definitely. I I definitely agree. I think you're totally right. Like um there are these things are quite good at sort of the the well-trodden paths of of the current and past reality. And sometimes you can take the like currently well-trodden paths and like project them out into the future, and they're probably going to be decent at that. But like when things are totally new and the world is changing, it's very actually things are some things are totally unpredictable by LLMs, but are um I think humans are still still pretty good at um sometimes figuring that out or making it happen. It's not just like um figuring it out. It's like just having the force of will or the vision to be like this can work, and I want to do it. Yeah, and I I think one of the other, you know, reasons I'm doing ChatGPT other than it's like a utility to me, it's very useful, is I do think product leaders and product managers and engineers, if you're not thinking about how to build actual like end user products in this new model with these new technologies, you're going to just be far behind in terms of what that future is going to look and feel like to users. So, I'll just like I'll give an example of this, which is we had I had this onboarding for ChatGPT. And it was fine. It was like three steps, and it was like fill out your profile, um do something else and like billing. And it was fine, and I just but I looked at it, and I was like this just doesn't feel like an onboarding to an AI product, it feels like an onboarding to like any other SAS SAS product. It just feels the same. And so, one of the things I started to play with, which is this um new onboarding that we're doing, is it was like it's a chatbot. So, like shouldn't the chatbot onboard you Yeah. and tell you like what's going on here? And so, it's okay. Like Capyera is telling me what's going to happen. Okay, then I get this form. Okay, we we haven't done the form yet, but we'll we'll get there. And then um it's like, "Great, I'm nice to meet you. Here's all the places you can find us. Here's all the things like if you need to get help." And then it goes into into our plan. Um so, I just think that's been something that I've worked on and just try to play with you know, how is the how how are these how are these products going to like look and feel and should you do these like chat style onboardings like fake? It's like just like little transitions in, but it anchors you into what's the modality of this product going to be and how like how are you going to like go into it and experience it moving forward? And it's been opportunities to play with ideas like that that I think are really useful for me as a product person regardless of what happens with Capyera. Yeah, that makes that makes all the sense. So, I know you have I know you have some personal stuff that you want to share about how you use use AI, but before we get to that, is there anything else on the work side that that you want to show demo? Um no, I mean I think I I think I'll give one more story and you can decide to use it or not, but just going back to the the strategy side. Um this is really practical and very efficient for me, so Um one I our office is in Oakland, so I have to cross the Bay Bridge to get to the office. Um when people ask, "How do you do it all?" This is how I do it. So, I have 20 It's 23-minute commute. I've been at my company now for about 5 and 1/2 months. So, hitting you know, it's time for me to like deliver the like here's this the capital P capital S strategy like I thought about it. Here you go. Um and I have all the contacts and information in my head for what we should do. I think I have a clear vision. And like I'm going to have to write this thing down. And so, what I did is I opened up voice chat GPT. Um I haven't done chat uh voice for PRD yet, but it's on the road map. And I just said this is what I think the strategy should be and it was meandering. It was not well organized. And I said, "Oh, I'm thinking about this competitor and that competitor and this thing and that thing and blah blah blah." And at the end of my 22 minute commute, I had a very I think a very good articulation of like what I think our strategy was that I did with like my hands on the wheel crossing the Bay Bridge looking at the beautiful skyline. And so, I just I'm really leaning in because it it helps me do my work at every level just so much better. I have the same experience. I love it for that. Like I was recently negotiating a deal and we were sort of like pretty close to the end of it and I I had this like vague feeling that something about the deal needed to change, but I couldn't figure out why and I couldn't express it. And I just like took a walk and just like talked into chat GPT for like um like 30 minutes, but yeah, it's like all these disconnected things where it's like I just I can't put my finger on what it is, but there's something here that I can't say. And then at the end of it I was just like, "Can you summarize it?" And it just like laid out in three bullet points like the exact things that I was trying to say that I couldn't say. And I was like, "Yeah, of course of course that's what it is, you know?" And it's it's so helpful, you know? Yeah. Yeah. Yeah, exactly. Um so, let's let's I want to see some of the things that you you're using this in your personal life. You're using it to uh to to do things with your family. Like tell me about like tell me about what you're using it for. Um so when you have a calendar like mine and jobs like mine and a life like mine, you are just drowning in information and you're drowning in requests. And so one of the things I actually did really early on is automate my email in in two specific categories using open AI APIs where I felt like I was falling really far behind but I had a very high volume of of demands. And so those two categories were one being recruited for jobs and two stuff from my kids' school. And so I every day I'm this isn't like a flex but like every day you get this message of like, "Hey, we have the highest growth CPO job on the face of the earth. We're making X million dollars blah blah." And all of them are a no and in seven years if I want a job, I should like reply to these recruiters and be very polite and like tell them exactly what I'm looking for so that you know, whenever I'm looking for a job in some like long-term future, they know what I want. And so I built an automation that looks for emails that are from recruiters. It checks the email against I'm not taking a job but it checks the email against types of jobs I would take and then drafts an email back saying, "I'm not interested but these are the kinds of jobs I would take." And like recommends to me recommends to me sends me a Slack message that recommends to me next steps whether it's like follow up, ignore this person, or whatever. Um so that's been really useful and I found that very constructive and it helped me like just get on top of stuff especially when I was job seeking. How did you make that? Like what what's what's the thing that's reading your emails? Uh it's a Zap. So it's a Zapier thing. So what I do is you know, there's a dozen recruiting firms. So I look for everything with their email addresses, plus anything that says like recruiter or talent in it. I tag it automatically with a filter in Gmail. And then the Zap reads all new emails on that tag. And then sends the email to a completion endpoint in open AI and says like, "Answer this question about it. Like what should I do? What should the email be?" And then sends that answer back to me and then sends it to Slack and drafts an email. That's the best. And can can you send the email from Slack and you be like, "Yes, send it." Or do you have to go into Gmail and do the same thing? No, it goes it goes into a draft and then it just like tells me like you should go send this email. Yeah, yeah, yeah. That's amazing. I love it. And did you Is that how you got your current job? Oh, no. I'm going to give uh I'm going to I'm just going to shout out everybody. Uh no, that's not how I got my current job. The way I got my current job is CJ at Artisan Talent, who always has the best jobs, texted me cuz he knows And like that's the kind of stuff that AI is not going to do yet, is like texts me and said, "I know what job you're going to take." Very accurate. Uh but no, he did not he did not do that. Um so, that that's an example one, but the second place that I'm truly drowning in, any parent can empathize with this, is I get 7,000 emails a day from school. Like everything from fourth grade to this art project to like, "Your kid's bleeding in the street and you need to come pick him up." Like it all goes into this like they all have these apps. These apps have these messaging things. They send email notifications. They send push notifications. Um they're very long like these messages are very long and it's just it's so much content and I was losing like, "What do I need to pay attention to and what do I not?" So, I'll try to show my screen without doxing my kids. I think I found um a specific Yeah, I I found a good one. Okay. So, what this does is is reads the email Uh-huh. from schools. And then at 3:00 or 4:00 p.m. every day, cuz that's like when I need to pay attention. It's like before I pick up the kids. Yeah. It says, "Summarize these in the like most succinct way possible. Only bubble up stuff that's relevant to Henry and Theo, my kids. And then like tell me if I need to do anything." And so this is one day of emails, mind you. This is how much stuff. This is one My kid fell and like had glass in his hand and it was a whole whole thing and I had to email them and say like be gentle. Another thing is like I didn't do his costume, so I had to tell his teacher he was in the ER all night, I didn't do the costume. Um the other thing was the teacher saying no big deal, he's reading a book. Then there was like an all school sleepover that was postponed. And then like this is the one that I would have missed, truly as a working mom, which is like picture day is tomorrow. Like make sure you bring dragon muffins. Um and so it's this kind of like every day I look at this and I'm like, "Great. I know exactly what to do." This is probably like 3,000 words worth of email. That's And I'm just getting like the very succinct um summary of it. And so this has been like just truly a lifesaver for me. And I operate I mean everybody operates in Slack. I operate in Slack. You cannot email me things. You cannot push it into an app. I need it like where I do my work. Um so that has been that's been the lifesaver. And I you know, I think I'm pretty popular in the parents WhatsApp group, cuz whenever anybody's like, "Hey, what day is the or when am I supposed to?" I'm like screenshot my little AI bot and say, "I I got you." That's incredibly cool. I mean, that seems like a second product. Like I feel like a lot of parents would pay for that. Yeah, maybe. I'm like as a public good, I'm willing to just teach anybody how to I try to do things like I try to filter out things. Um you know, I try to be like pretty protective of of information, but like this stuff where it's like my kids, I got I got to get these summaries, Anne. Yeah. Yeah, that makes sense. That's that's truly incredible. I guess I'm I'm curious like sort of toward towards the end of this discussion, we've seen this incredible product you're building. Like what is the future of that for you? Like are you do you want to run this as a business where this is the only thing you do? Is it like continually a side project? Yeah, like how are you thinking about it? I want to see if we can get it to a million dollars. Mhm. That's like Where is it now? Can you share? What? Can you share what it's at? We're in six six figures of of ARR six months. Yeah. Side project. That's that's great. My My goal, I mean, I told everybody I was like my goal is I want to buy myself like a nice glass of wine once a week. Remember, I was selling this for like two bucks a pop. I was like, I get a nice glass of wine like once a month or once a week, I'll be happy. And then I was like, I might get some wine. And then my new ambition was I can pay for my kids' school. So I think we've we've definitely crossed that threshold. And now I'm like, can I pay for my kids' college? Like it's really kids It's kid-centric milestones. We're funding that 529. But I I think like intellectually, no, I Here's Here's what is very interesting about doing this is Look, I've got friends. Like would they write me a check? Could they fund this? Sure. Like you know where we met. Like yeah, people would probably write me a check. I could like knock on your door and be like, pretty please. But I'll write you a check if you want. Thanks. But my question That's not my question. My question is like does this kind of product need that kind of check? Yeah. And if it doesn't how do you build it like what's the What is this business? This is like more the intellectual side of me, which is I'm actually trying to as somebody who really cares and loves technology and somebody who loves startups, I don't know if the current model of venture funding is the right model for something like this. Maybe it is, maybe it's not. I just don't think it's not it's not obvious to me that it is. And so what I'm trying to do is push the edges of how far you can take this with me and Alyssa, who works part-time. And look, if I hit a wall where capital's the challenge, like I can get access to capital, but right now it's not times the challenge. It's There's there's something else here, which I'm trying to figure out how to solve the right problem, how to access the right people. So we just I I'm very happy at my job. I love Lunch Darkly. It is a product that I'm just like so happy to work on. I'm an engineer. It just it like gives me a lot of joy. I'm pretty good at the job that I do. They know about Chat Period, they're extremely supportive of it. Um and we've carved out the IP cuz we're all good corporate citizens. Um but I this is this is a this is it's an exercise and an a experiment for me, and I want to see how it plays plays out. Like never say never, right? If 10,000 users turns to 100,000 users in the next 6 months, like we might have a different conversation. Um but let's let's see how it goes. Yeah, I think I think that that makes a lot of sense, and and I I love this sort of point about um uh new funding models for these kinds of companies. Like the venture model was built for like when software was a lot riskier to build. Um and and the outcomes were also like much much much bigger potentially. Um and there's a lot of room for these sort of AI businesses where it's like uh it can actually be a really meaningful like cash-generating business, but it's probably not going to be like $10 million or something like that. Um and it's something that like thought about a lot at Every cuz we have a media business, um but we've also we've already spun out an AI writing app called Lex. Um and we have a couple more incubations and like the amount of leverage that we get with like we have six people full-time now, ish, and like the amount of leverage we get with using AI tools like we're shipping so much stuff all the time and it's like it's like a million-dollar year business or something like that, but like if we had a little bit more capital like I think we could do like amazing stuff, but I just don't want to raise another actual venture round and so it's like what is the shape of that money and and and all that kind of stuff is really I think an important question. Well, and you and I know this, but you know, there was I was at an event and somebody asked the question of like, when do you know it's the right time to raise venture capital? And there was like a lot of interesting answers about product market fit. And my answer was like, one, when you know you can do something with it, and two, when you want a venture outcome. And a hundred million dollars is not a venture outcome. Like people don't understand that. Um A hundred million dollars would change my life. I don't know about you. It would change my life. I'd take it. I would have a fund of 529. So, I think you really do have to think about like you can be extremely ambitious. I don't think a hundred million dollars is something to sneeze at. It's not a venture outcome and you're you start playing the venture game, that is the the path you commit yourself to to your investors. I know that. I've taken venture capital before. Um so, I just think that, you know, also this is like not my first round. Uh and so I have a lot more flexibility. I can self-fund stuff if I need to even though we're um obviously like profitable relative to kind of like time and and investment. So, I think it's interesting. I have a last thing to show you if you want to see it. I would love to see it. Bring it on. Okay, these are fun things. Which is I'm raising the next generation of like AI AI-enabled children. So, I built three things I built three things with my kids um using AI. So again, this is like less about the like logistical how do I keep my life on top of school stuff and more about like cool ways you can introduce kids to AI. So I'll show you like just a couple things that I've built that I I'm proud of. I think they're very fun and I did them with my my 6-year-old. So the first one Okay, so the first one is this, which is my then four and six-year-old at the time, now five and seven-year-old at the time, and I they're really into Greek mythology. And so we built these Pokémon-style cards including images for them to like fake play Greek mythology Pokémon. And like stuff with it. It was so fun. They're obsessed with these. They keep them in their like little file folder with all their all their Pokémon cards. So this was just like a really fun art project. I guess two of these are Greek mythology-related. And then I they're also into this podcast called Greeking Out, which I highly recommend. My kids really into it, so I made him a little podcast hub of podcasts that he likes. And it's the podcast episode, let's see if it's work. I haven't even spun this up in a long time, so we'll see. So basically what I built is a podcast hub that takes all these podcasts he likes, transcribes them, and gives little quizzes at the end. So he can like listen to stuff about Poseidon and then like do a question and answer with this app. It's apparently not running right now. But but that's a really fun one. And then the last one, this was our Christmas break project, is we built runawaypancakes.com. And so one of the problems I have as a parent is recipes are not built for kids to follow because it has the big list of like instructions at the top, and then all these instructions you have to go back and it's like add the flour to the water." And it's like, "How much flour? How much water?" So, um my 7-year-old and I built out his favorite recipes um as and we generated all the mid-journey image These are all mid-journey images. Most of them are not creepy. Um but we built all his favorite recipes, but we did them in a way where he could follow them step-by-step. And it had illustrations with um kind of like pictures for what it might look like that we did out of mid-journey. And so, this was a really fun project. So, you have the ingredients here. And what we did is we took the recipes and then each step This is like the big innovation. It was like, "Mix this much flour, this much baking powder, this much sugar." Cuz this is not what um recipe sites tend to be to be doing. And so, you can see you can go through step-by-step. You can kind of like see what it looks like. Um and then you get your fluffy pancakes. And then we made a whole bunch of coloring pages. That's so cool. This is the best. Like being a kid right now seems so cool because like yeah, it's yeah, like recipes are not built for for kids. Like I had I ran into that so much growing up. It's like, "I'm interested in this. I want to make food. I want to learn how to write. I want to do all this stuff." And it's like, "There weren't classes for like 7-year-olds cuz there aren't that many 7-year-olds that like want to do that." But like AI stuff means like uh you and your parent can just like make something like this. And it's I just it's delightful. I'm It makes me so happy. My uh maybe chaotic good thing that I did is I trained a 11 Labs voice on like Santa. And then when my 4-year-old was being great in the December time period Santa left him a voicemail. It was like, "Theo, I've heard you've been a little disrespectful to your mom during dinner." Did it work? It would be like I'd be like, "Do you want to talk to Santa?" And he'd be like, "No." And I'd be like, "Santa has something to say." And then I'm like So, there's lots of lots of little hacks you can do with it as a A lot of good parenting tips in this episode. I like it. Good dad take them for what they're they're worth. But now my kids demand custom stories from me. They're like, "I want a story about a like wild chicken who becomes a prince on a island kingdom and then fights Darth Vader." And you're like, "Okay, dude. Like I got I'm like on it." You know? So, yes, in one ways it's very magical in other ways like spoiled for choice. Yeah. Yeah, I wonder I I just wonder what it's like to to grow up in a world where all your questions are answerable and all your stories are custom. Like it's kind of a different it's a different thing, you know? Oh, yeah. I think I closed it. One of the chats was like, "My son was convinced there was a bucket war in ancient Italy." Hm. He was like, "Mom, when was the bucket war?" And I'm like, "What the hell are you talking about?" My husband and I are like pretty into history and we're like, "Bucket war?" We're googling, like, "What is the bucket war?" And we asked um ChatGPT. We said something like, "What was the bucket war in Florence?" Or something. And it was like, "Oh, you're mixing up two historical events. This event and that event." And there was actually a war over a bucket um in ancient Italy. Um I'll try to find the link. We can put it in the show notes. Please, we'll put it in the show notes. Um Being able to interpret my children's very vague recollection of almost accurate facts is a is a very useful use case. Incredible. I love it. Um this has been so fun. Thank you so much for doing this uh with me uh and for sharing everything. Um I love what you're up to. Um just for just as we end like if people want to know more, if they want to uh message you, follow you on social, use chat chat PRD, like where can they find you? Yep, so I'm on X Clarvo, all one word. I'm on TikTok, Chief Product Officer, if you really want some very niche product content, that's the place to go. It's a little stale. I've been a little busy the last 6 months, but we'll we'll revive it. Chatprd.ai is where we're at and Chatprd is on Twitter, LinkedIn, and YouTube. Very brand new to I'm trying I'm trying to learn YouTube. So, that's my that's my next project. Amazing. Um, well, thank you. Oh my gosh, thank you. This was fun. Oh my gosh, folks, you absolutely positively have to smash that like button and subscribe to AI and I. Why? Because this show is the epitome of awesomeness. It's like finding a treasure chest in your backyard, but instead of gold, it's filled with pure unadulterated knowledge bombs about ChatGPT. Every episode is a roller coaster of emotions, insights, and laughter that will leave you on the edge of your seat craving for more. It's not just a show, it's a journey into the future with Dan Shipper as the captain of the spaceship. So, do yourself a favor. Hit like, smash subscribe, and strap in for the ride of your life. And now, without any further ado, let me just say Dan, I'm absolutely, hopelessly in love with you.