Podcast Summary: How I AI
Episode: How to turn meeting notes into prototypes your sales team can immediately demo to customers
Guest: Anjan Panneer Selvam, CPTO of Acolyte Health
Host: Claire Vo
Date: September 1, 2025
Episode Overview
This episode centers on using AI tools to transform raw meeting notes—often from high-stakes executive or stakeholder meetings—into functional, interactive software prototypes that sales and customer success teams can immediately demo to customers. Anjan Panneer Selvam discusses in detail his step-by-step workflow, the culture shifts enabled by this new way of building, and the tangible impact on speed, alignment, and team collaboration. The discussion features demos, anecdotes, and practical advice for product leaders, engineers, and startup executives.
Key Discussion Points & Insights
1. The CPTO Role & AI’s Impact on Organizational Structure
-
Blending Product and Engineering:
- Anjan explains how AI blurs the lines between CTO and CPO, leading to the role of CPTO, particularly valuable in early-stage startups.
- Quote at [03:19]:
“AI definitely has made it much more bearable and easier to manage the responsibilities that come with being both responsible for the product and technology side of things.” — Anjan
-
Reducing Information Transfer Friction:
- AI accelerates the transition from idea to execution, minimizing the need for traditional handoff and translating processes.
2. Workflow: From Meeting to Prototype in Under an Hour
-
Capturing Stakeholder Input:
- Anjan records meetings (sometimes using the Limitless Pendant wearable) to capture organic discussions with stakeholders or the CEO.
- Those transcripts are then pasted into ChatGPT for initial ideation and prompt generation.
-
From Transcript to Prompt:
- The team skips the “structured PRD” and uses a natural-language brain dump to ChatGPT, which helps consolidate and translate ideas into structured prompts.
- Memorable Quote at [07:26]:
"AI helps normalize every good product manager’s product sense...into very structured information. AI doesn’t judge you." — Anjan
-
Prototype Generation:
- Prompts, sometimes enhanced with technical references (e.g., React Flow), are fed into tools like Lovable, V0, or Magic Patterns to auto-generate interactive prototypes.
-
Hardware vs. Software Note Capture:
- The Limitless Pendant offers non-intrusive, multilingual, spontaneous transcription, helping to lower friction for capturing ideas in any setting.
- Anjan at [09:39]:
“There should be no friction. Oh, let me take a paper and write it down...this has completely changed the game.”
3. Rapid Alignment with Visual Prototypes
-
Immediate Stakeholder Feedback:
- Interactive prototypes radically reduce alignment time across product, engineering, and executive leadership—from weeks to hours.
- These prototypes replace lengthy requirement documents and static Figma files with clickable, editable prototypes.
-
Reducing Missed Expectations:
- The “cheap path to failure”: If a demo is off the mark, you’ve lost 15 minutes, not weeks.
- Claire at [15:53]:
“The worst thing is you say, okay, I’ll come back in two or three weeks and show it to you, and you show it and it’s totally wrong... Now, if it’s not what [the CEO] wanted, it cost you...15 minutes.”
4. Enabling Research, Slides, & Analysis in Minutes
-
Workflow Efficiency:
- Using Gamma and Perplexity AI for deep quick-dive market research, customer persona identification, and competitive analysis.
- Exporting results for decks/slides in minutes rather than days.
-
Pro/Con and “Devil’s Advocate” Analysis:
- Anjan prompts AI (“play devil’s advocate”) to temper its “abundance mindset” and surface potential pitfalls, not just opportunities.
- Quote at [23:38]:
“There are more features that AI has helped me say no to because it’s all in front of me...you can more freely say no and only focus on things that are important.” — Anjan
5. PRDs, Validation, and Internal “Gates”
-
Chat PRD Integration:
- Anjan uses Chat PRD for structured requirements validation, customized to his team’s preferred format, ensuring all critical questions are answered before engineering invests time.
- Chat PRD also extracts V2/”future enhancements” separately to tease further iteration.
-
AI as an Internal Reviewer:
- Claire at [27:38]:
“Using AI...as a gate to make sure you’ve checked all the boxes...is a really effective way to bring AI into the loop [and] save time.”
- Claire at [27:38]:
6. Democratizing Prototypes for Customer & Sales Team Access
-
The Live Demo Library:
- All prototypes, even work-in-progress, are published as a “living library” of interactive demos.
- Sales and customer success teams can access and share these with customers for instant feedback, moving the product from slideware to interactive experiences—sometimes within 30 minutes of initial ideation.
- Anjan at [29:30]:
“Sales teams or customer success teams...have now the flexibility to take this and show it to customers, get real time feedback.”
-
Changing Product-Sales Collaboration Culture:
- Moves away from secrecy; instead, teams are open about prototypes being “not live yet”, improving mutual trust and transparency.
7. Deconflicting and Breaking Deadlocks with AI
-
Unblocking Engineering with “It’s Possible” Prototypes:
- When engineering is resistant (“we have no mobile developer!”), a proof-of-concept mobile app is generated with Roark or similar tools within 10 minutes.
- The intention isn’t to pressure engineers but to show new possibilities and inspire discussion, not present a “fait accompli.”
-
Changing Team Culture & Org “Rules”:
- Claire and Anjan reflect on how this workflow reshapes traditional power dynamics and expectations across teams.
- Claire at [39:00]:
“There are sort of these, like, artificial rules and blocks that we...have put in place. [Now] there’s just so much more access to overcome those misunderstandings, get more creative, be more inspired.”
8. Looking Forward: How AI Will Change Company Workflows
-
Meta Predictions:
- Companies will spend more time validating up front, align faster across departments (including customer support), and reduce waste from confusion after shipping.
-
The “Break” When AI Fails:
- Not every AI prototype works; patience and iteration are necessary. Sometimes you just need to “walk away and take a break.”
- Quote at [45:51]:
“Sometimes you just need to take a break and stop prompting until end of credits. For the one demo I showed, I have at least three that did not work.”
Notable Quotes & Memorable Moments
-
On the Limitless Pendant:
- “There should be no friction...this has completely changed the game.”
— Anjan, [09:39]
- “There should be no friction...this has completely changed the game.”
-
On Rapid Prototyping:
- “At the end of this, you know, I have a fully functional...drag and drop canvas builder that could build a whole user journey.”
— Anjan, [13:12]
- “At the end of this, you know, I have a fully functional...drag and drop canvas builder that could build a whole user journey.”
-
On Risk-Free Iteration:
- “If you get this and CEO is like, no, it’s not what I want, it cost you what, 15 minutes. Not a big deal.”
— Claire, [15:53]
- “If you get this and CEO is like, no, it’s not what I want, it cost you what, 15 minutes. Not a big deal.”
-
On AI’s Abundance Mindset:
- “AI has an abundance mindset...I always have a follow up saying, can you play devil’s advocate?”
— Anjan, [23:38]
- “AI has an abundance mindset...I always have a follow up saying, can you play devil’s advocate?”
-
On Enabling Sales & Customer Teams:
- “Thirty minutes later, we’re with the customer actually demoing this because they wanted to see it.”
— Anjan, [29:30]
- “Thirty minutes later, we’re with the customer actually demoing this because they wanted to see it.”
-
On Team Culture Shifts:
- “For the longest time, what was friction could easily now be converted to a partnership.”
— Anjan, [33:53]
- “For the longest time, what was friction could easily now be converted to a partnership.”
-
On Breaking Deadlocks:
- “The goal here is not to say something is easy. It’s more, are we all aligned and can we move fast?”
— Anjan, [41:55]
- “The goal here is not to say something is easy. It’s more, are we all aligned and can we move fast?”
-
On AI Limitations:
- “When AI doesn’t work, you take a step back and use something else, more broad, and be patient.”
— Anjan, [45:51]
- “When AI doesn’t work, you take a step back and use something else, more broad, and be patient.”
Timestamps for Key Segments
| Timestamp | Segment/Topic | |-----------|---------------------------------------------------------------------------------------------| | 02:39 | The CPTO role and the impact of AI on org structure | | 04:47 | Workflow: Capturing meeting notes and generating prototypes | | 09:39 | The Limitless Pendant: hardware for seamless transcription and ideation capture | | 13:12 | Interactive prototypes in Lovable—reducing alignment time | | 19:24 | Using AI tools for market research, slides, and analysis | | 23:38 | Forcing AI to play "devil’s advocate" and not just provide positive feedback | | 25:23 | Using Chat PRD for requirements validation; "gating" before engineering | | 29:30 | Giving sales and CS teams instant access to live demo libraries | | 36:15 | Breaking deadlocks with AI-powered mobile app prototypes (e.g., with Roark) | | 39:00 | Org culture: Redefining traditional product/engineering/sales dynamics | | 43:36 | How AI and prototyping will reshape company processes | | 45:51 | What to do when AI gives poor output; the value of patience and iteration |
Takeaways & Practical Tips
- Use meetings as data: Always record (with consent) and transcribe stakeholder discussions for raw, unstructured material.
- Leverage AI for structure: Dump the transcript into ChatGPT and similar tools to generate structured, actionable prompts.
- Prototype rapidly: Feed those prompts into prototyping tools (Lovable, V0, Magic Patterns) for an interactive demo within minutes.
- Validate at each step: Use targeted AI prompts for market research, PM documentation, and devil’s advocate analysis.
- Democratize demos: Maintain an open, living library of prototypes for internal and customer-facing teams.
- Foster a culture of alignment: The goal isn't to eliminate roles, but to maximize speed, creativity, and trust across org boundaries.
- Iterate—don’t expect perfection: Not every prototype will hit the mark; take breaks and use different tools when needed.
Conclusion & Speaker Contact
Anjan emphasizes a cooperative, practical approach to integrating AI in product workflows—moving from alignment to rapid validation, all without burdening engineering. The tools and mindset discussed represent a shift not only in productivity but in company culture.
Contact:
- LinkedIn (preferred), Twitter/X
- Open to conversation about advocating for more AI in product/tech workflows
For more episodes and resources, visit howiaipod.com
