How Lattice’s PM team built an AI Slack “Intern” | Neha Monga, CPO (Ex-Lattice) | LaunchPod
AI isn’t a “nice to have” for PMs anymore, it’s a baseline.
TL;DR:
On today’s episode of LaunchPod, Neha Monga, former Head of Product at Lattice and product leader at Meta, Amazon, and Expedia, joined LaunchPod to share the exact AI workflows her teams use to move faster, validate smarter, and stay ahead.
We break down:
The AI workflows that are table stakes for PMs now, and how to adopt them so you don’t fall behind (02:38)
How Neha’s team used Figma Make + Claude to spin up high-fidelity prototypes, gather customer feedback fast, and slash validation time with design and engineering (11:09)
Lattice’s process behind building their PM “intern” inside Slack to instantly surface insights from their massive customer feedback library (18:55)
What do you want to learn about when it comes to incorporating AI into your team’s workflow?
1. Adopt these AI workflows or fall behind
If you’re still taking manual meeting notes, writing status reports from scratch, or digging through feedback by hand, Neha has one message for you: You’re falling behind.
“The best way to learn AI is actually by using it and seeing how other products are being built.”
Her team used tools like:
Zoom AI or Granola to automatically record and summarize meetings.
ChatGPT/Gemini to refine decision docs and status updates.
Gmail’s built-in AI features to speed up communication and synthesis.
Pro tip: Even without full enterprise support, Neha recommends lightweight tools like Granola for high-quality personal meeting summaries, especially when you’re doing advisory or side work.
2. Prototype → Feedback → Iterate: In hours, not weeks
Neha’s team slashed validation time by combining Claude and Figma Make to create high-fidelity prototypes before designers or engineers ever touched a ticket.
“Within three or four prompts, I can have something I can put in front of customer and a picture stays a thousand words and they can understand… I think that is magical being able to do this.”
How they used it:
Rapidly mocked up features with Figma Make using real design systems.
Plugged those into Claude to generate UX flows or refine interactions.
Brought customer feedback in early, before anything hit a backlog.
This not only accelerated customer validation but also improved internal communication. “It saves hours even just explaining things to design or engineering,” she said. “Seeing is so much faster than describing.”
3. The Slack-based “PM Intern” that changed everything
One of the most powerful tools her team built? A custom GPT-powered PM intern that lived inside Slack.
Here’s how it worked:
Hooked up to a folder of structured customer feedback (from Salesforce, Gong, support tickets, etc.)
Used a custom GPT trained only on internal data, not the open web.
Answered natural language queries like:
“What are our top analytics requests from SMB customers?”
“Who else has asked for this feature?”
“What’s trending in customer feedback on engagement?”
“This was less than half an hour to set this up. The best part is it's not just a product team, right? Anyone can use it. So it's a marketer or a salesperson trying to go into a customer meeting and they wanna pull out a few things. That has been just a game changer.
The intern turned what used to be multi-day, cross-system research projects into Slack messages with instant answers. And critically, it democratized feedback access across the company, not just for product.
Links
Tools
Chapters
What does LogRocket do?
LogRocket's Galileo AI watches user sessions for you and surfaces the technical and usability issues holding back your web and mobile apps. Understand where your users are struggling by trying it for free at LogRocket.com.