How I AI

• How Amplitude built an internal AI tool that the whole company’s obsessed with (and how you can too) | Wade Chambers

  Wade Chambers, Chief Engineering Officer at Amplitude, shares how his team built Moda—an internal AI tool that gives employees access to enterprise data across multiple systems, enabling faster product development and decision-making while fostering cross-functional collaboration.What you’ll learn:1. How Amplitude built a powerful internal AI tool in just 3 to 4 weeks of engineers’ spare time2. A social engineering approach that made their AI tool go viral company-wide in just one week3. How product managers use AI to analyze customer feedback across multiple data sources and identify key themes4. A streamlined workflow that compresses research, PRD creation, and prototyping into a single meeting5. Why role-swapping exercises with AI tools build empathy and cross-functional fluency across product, design, and engineering teams6. How AI tools are helping engineering teams tackle persistent tech debt challenges more effectively—Brought to you by:CodeRabbit—Cut code review time and bugs in half. Instantly.Vanta—Automate compliance and simplify security—25k giveaway: To celebrate 25,000 YouTube followers, we’re doing a giveaway. Win a free year of my favorite AI products, including v0, Replit, Lovable, Bolt, Cursor, and, of course, ChatPRD, by leaving a rating and review on your favorite podcast app and subscribing to the podcast on YouTube. To enter: https://www.howiaipod.com/giveaway.—Where to find Wade Chambers:LinkedIn: https://www.linkedin.com/in/wadechambers/Amplitude: https://amplitude.com/blog/meet-the-team-wade-chambers—Where to find Claire Vo:ChatPRD: https://www.chatprd.ai/Website: https://clairevo.com/LinkedIn: https://www.linkedin.com/in/clairevo/X: https://x.com/clairevo—In this episode, we cover:(00:00) Introduction to Wade Chambers(02:53) The build vs. buy decision for internal AI tools(04:55) What Moda is and how it works(07:19) The social engineering approach to adoption(09:17) Demo of Moda in Slack(10:58) Data sources Moda has access to(12:43) Analyzing customer feedback themes with Moda(17:41) Behind the scenes: how Moda works technically(23:24) Creating a PRD from a single customer insight(27:30) How teams actually use AI-generated PRDs(29:09) Impact on product development velocity(32:37) Engineers, designers, and PMs swapping roles(34:38) Recap of creating Moda(36:00) Lightning round and final thoughts—Tools referenced:• Glean: https://www.glean.com/• ChatGPT: https://chat.openai.com/• Cursor: https://cursor.com/• Bolt: https://bolt.new/• Figma: https://www.figma.com/• Lovable: https://lovable.dev/• v0: https://v0.dev/—Other references:• Amplitude: https://amplitude.com/• Slack: https://slack.com/• Confluence: https://www.atlassian.com/software/confluence• Jira: https://www.atlassian.com/software/jira• Salesforce: https://www.salesforce.com/• Zendesk: https://www.zendesk.com/• Google Drive: https://drive.google.com/• Productboard: https://www.productboard.com/• Zoom: https://zoom.us/• Asana: https://asana.com/• Dropbox: https://www.dropbox.com/• GitHub: https://github.com/• HubSpot: https://www.hubspot.com/• Abnormal Security: https://abnormalsecurity.com/—Production and marketing by https://penname.co/. For inquiries about sponsoring the podcast, email [email protected].

  --------  

  40:26

  --------

  40:26
• An exclusive inside look at GPT-5

  In this episode, I share my hands-on experience with OpenAI’s GPT-5, the company’s new frontier model. As one of the first users outside of OpenAI to test the model, I put GPT-5 head-to-head with GPT-4.1 across real-world product use cases—from writing PRDs to generating code to assisting with visual design work. This is my unfiltered look at what GPT-5 can (and can’t) do—and how it changes the game for builders.What you’ll learn:1. How GPT-5 differs from previous models with its engineering-focused approach to problem-solving and tendency to prioritize technical details over business context2. A comparative analysis of how GPT-5 and GPT-4.1 generate different types of product requirement documents and prototypes for the same prompt3. Why GPT-5 excels at technical writing, functional requirements, and code generation while potentially skipping important business discovery questions4. The model’s impressive spatial awareness capabilities when generating images for interior design and other visual tasks5. Practical considerations for choosing the right model based on your specific use case and audience6. How GPT-5’s extensive tool-calling behavior and bullet-point communication style reflect its engineering-oriented design—Brought to you by ChatPRD—an AI copilot for PMs and their teams: https://www.chatprd.ai/howiai—25k giveaway: To celebrate 25,000 YouTube followers, we’re doing a giveaway. Win a free year of my favorite AI products, including v0, Replit, Lovable, Bolt, Cursor, and, of course, ChatPRD, by leaving a rating and review on your favorite podcast app and subscribing to the podcast on YouTube. To enter: https://www.howiaipod.com/giveaway—Where to find Claire Vo:ChatPRD: https://www.chatprd.ai/Website: https://clairevo.com/LinkedIn: https://www.linkedin.com/in/clairevo/X: https://x.com/clairevo—In this episode, we cover:(00:00) Introduction to GPT-5(04:34) Testing GPT-5 in ChatPRD for document generation(07:10) Comparing GPT-5 and GPT-4.1 on business vs. technical orientation(11:22) Side-by-side comparison of PRDs generated by both models(15:23) Where GPT-5 excels: Technical considerations and documentation quality(17:35) Comparing prototypes generated from different model outputs(19:57) Testing homepage critique capabilities between models(23:14) OpenAI’s strengths in API design and developer support(25:37) GPT-5’s performance as a coding assistant(27:26) Examining GPT-5 in ChatGPT’s interface(28:50) Testing GPT-5’s front-end design capabilities(31:17) Personal use case: bathroom remodel planning(33:45) Comparing GPT-5 vs. GPT-4 for interior design visualization(38:10) Summary of key findings and recommendations—Tools referenced:• OpenAI: https://openai.com/• ChatGPT: https://chat.openai.com/• Claude: https://claude.ai/• Gemini: https://gemini.google.com/• Cursor: https://cursor.sh/• v0: https://v0.dev/• Lovable: https://lovable.dev/• Bolt: https://bolt.com/• LaunchDarkly AI Configs: https://launchdarkly.com/docs/home/ai-configs—Other reference:• Benjamin Moore paints: https://www.benjaminmoore.com/—Production and marketing by https://penname.co/. For inquiries about sponsoring the podcast, email [email protected].

  --------  

  40:11

  --------

  40:11
• How a VC and tech founder used AI to launch a brick-and-mortar business in their spare time | Andrew Mason (CEO of Descript) & Nabeel Hyatt (Partner at Spark Capital)

  Andrew Mason (founder of Groupon, now CEO of Descript) and Nabeel Hyatt (General Partner at Spark Capital) teamed up to open a physical board-game social club in Berkeley, with AI as their business partner. In this episode, they break down how they used Claude to generate a full business plan, model financials, plan the space layout, navigate Berkeley permitting, categorize hundreds of games using a custom Dewey Decimal–style system, and build an AI concierge that matches players with games via text. They also share how working on this side project helped rewire how they use AI in their day jobs—and why more people should use AI to build real-world things.What you’ll learn:1. How to use Claude Projects as your business copilot to create comprehensive business plans, financial projections, and space layouts2. A workflow for categorizing hundreds of board games using an AI-generated “Dewey Decimal System” that makes game discovery intuitive3. How they built an AI concierge service that matches players with games and coordinates group play sessions via text message4. Why AI enables side projects that would otherwise be impossible due to time constraints and specialized knowledge requirements5. A simple system for creating customer personas that inform your business model and event programming6. How to use model context protocols (MCPs) to connect AI assistants to business tools like Airtable without complex coding—Brought to you by:Lovable—Build apps by simply chatting with AIPersona—Trusted identity verification for any use case—Where to find Andrew Mason:LinkedIn: https://www.linkedin.com/in/andrewmason/X: https://x.com/andrewmason—Where to find Nabeel Hyatt:LinkedIn: https://www.linkedin.com/in/nabeelhyatt/X: https://x.com/nabeel—Where to find Claire Vo:ChatPRD: https://www.chatprd.ai/Website: https://clairevo.com/LinkedIn: https://www.linkedin.com/in/clairevo/X: https://x.com/clairevo—In this episode, we cover:(00:00) Introduction to the board-game social club concept(02:44) How AI made a challenging side project possible(06:14) Using Claude as a business copilot for planning(12:53) Developing customer personas with AI(15:45) Using AI to determine business viability(21:02) Navigating Berkeley real estate and permitting(25:18) Building an AI concierge for game matchmaking(28:10) Database design with Airtable for non-technical founders(32:04) Creating a custom board-game categorization system(36:20) Demo of the text-based AI concierge service(40:38) Enabling experiences that wouldn’t exist without AI(43:42) Lightning round and final thoughts—Tools referenced:• Claude: https://claude.ai/• Airtable: https://airtable.com/• n8n: https://n8n.io/• Twilio: https://www.twilio.com/• Cursor: https://cursor.sh/• Windsurf: https://www.windsurf.io/• Python: https://www.python.org/—Other references:• Model context protocol (MCP): https://www.anthropic.com/news/model-context-protocol• Tabletop Library: https://tabletoplibrary.com/• Descript: https://www.descript.com/—Production and marketing by https://penname.co/. For inquiries about sponsoring the podcast, email [email protected].

  --------  

  48:54

  --------

  48:54
• How Block’s custom AI agent supercharges every team, from sales to data to engineering | Jackie Brosamer & Brad Axen

  VP of engineering Jackie Brosamer and principal engineer Brad Axen join me to demo Goose, Block’s open-source AI agent that runs locally, plugs into your existing tools through model context protocol (MCP) servers, and peels away the rote parts of work so people can focus on insight and impact.This episode is packed with in-depth demos: starting with a messy farm-stand sales CSV, Goose analyzes the data, builds visualizations, and generates a shareable HTML report. We then spin up an MCP that lets Goose talk to Square’s dashboard for inventory management, vibe code an email MCP that can send payment links automatically, and unpack how environment setup, debugging, and tool orchestration get handled behind the scenes.What you’ll learn:A practical, repeatable workflow for turning any working script or function into a custom MCP—and exposing it to natural-language controlHow to transform messy CSVs into visualizations, HTML reports, and actionable business insights without needing a data science backgroundWays to hook Goose into live business systems (e.g. Square inventory, payments) so analysis flows directly into operational actionThe thinking behind Block’s decision to open-source GooseLessons from Block’s bottom-up meets top-down adoption modelWhy organizational transformation, not just picking the right LLM, will separate AI winners from laggards over the next few yearsHow to scale an internal MCP catalogThe organizational transformation required to fully leverage AI capabilities—Brought to you by:CodeRabbit—Cut code review time and bugs in half. Instantly.Lenny’s List—Hands-on AI education curated by Lenny and Claire—Where to find Jackie Brosamer:LinkedIn: https://www.linkedin.com/in/jbrosamer/—Where to find Brad Axen:LinkedIn: https://www.linkedin.com/in/bradleyaxen/—Where to find Claire Vo:ChatPRD: https://www.chatprd.ai/Website: https://clairevo.com/LinkedIn: https://www.linkedin.com/in/clairevo/X: https://x.com/clairevo—In this episode, we cover:(00:00) Introduction to Goose and its data analysis capabilities(02:27) How Block embraced AI across the organization(04:48) What Goose is and why Block open-sourced it(07:45) Demo: Analyzing farm-stand sales data with Goose(12:18) Creating shareable HTML reports from data analysis(14:15) Model context protocols (MCPs) that Goose uses(18:56) Demo: Using Square MCP to create a product catalog(23:35) Creating payment links from analyzed data(26:30) Demo: Building a custom email MCP(31:18) Testing the new email MCP with Goose(36:09) Debugging and fixing MCP code errors(38:44) Connecting workflows: sending payment links via email(41:30) Lightning round and final thoughts—Tools referenced:• Goose: https://block.github.io/goose/• Pandas: https://pandas.pydata.org/• Plotly: https://plotly.com/• Python: https://www.python.org/• ChatGPT: https://chat.openai.com/• Claude: https://claude.ai/• Cursor: https://www.cursor.com/• Mailgun: https://www.mailgun.com/—Other references:• Block: https://block.com/• Model context protocol (MCP): https://www.anthropic.com/news/model-context-protocol• GitHub: https://github.com/—Production and marketing by https://penname.co/. For inquiries about sponsoring the podcast, email [email protected].

  --------  

  46:31

  --------

  46:31
• Successfully coding with AI in large enterprises: Centralized rules, workflows for tech debt, and training your team | Zach Davis (Director of Engineering at LaunchDarkly)

  Zach Davis is a product-minded engineering leader and builder at heart, with over 12 years of experience building high‑performing teams and crafting developer tools at companies like Atlassian and LaunchDarkly. In this episode, he shares how he’s helping his 100-plus-person engineering team successfully adopt AI tools by creating centralized documentation, using agents to tackle technical debt, and improving hiring processes—all while maintaining high quality standards in a mature codebase.What you’ll learn:1. How to create a centralized rules system that works across multiple AI tools instead of duplicating documentation2. A systematic approach to using AI agents like Devin and Cursor to analyze and reduce test noise in large codebases3. How to leverage AI tools to document your codebase more effectively by extracting knowledge from existing sources4. Why “what’s good for humans is also good for LLMs” should guide your documentation strategy5. A custom GPT workflow for improving interview feedback quality and coaching interviewers6. How to approach tech debt reduction with AI by creating prioritized task lists that both humans and AI agents can work from—Brought to you by:WorkOS—Make your app enterprise-ready todayLenny’s List on Maven—Hands-on AI education curated by Lenny and Claire—Where to find Zach Davis:LaunchDarkly: https://www.launchdarkly.comLinkedIn: https://www.linkedin.com/in/zach-davis-28207195/—Where to find Claire Vo:ChatPRD: https://www.chatprd.ai/Website: https://clairevo.com/LinkedIn: https://www.linkedin.com/in/clairevo/X: https://x.com/clairevo—In this episode, we cover:(00:00) Introduction to Zach Davis(02:44) Overview of AI tools used at LaunchDarkly(04:00) The importance of having someone responsible for driving AI adoption(05:44) Why vibe coding isn’t acceptable for enterprise development(06:42) Making engineers successful with AI on their first attempt(07:55) Creating centralized documentation for both humans and AI agents(10:19) Using feature flagging rules to improve AI outputs(12:33) Advice for getting started with rules(14:28) Demo: Setting up Devin’s environment in a large codebase(24:33) Devin’s plan overview(27:55) Demo: Creating a prioritized tech debt reduction plan(36:40) Demo: Using AI to improve hiring processes and interview feedback(40:34) Summary of key approaches for integrating AI into engineering workflows(42:08) Lightning round and final thoughts—Tools referenced:• Cursor: https://www.cursor.com/• Devin: https://devin.ai/• ChatGPT: https://chat.openai.com/• Claude: https://claude.ai/• Windsurf: https://windsurf.com/• Lovable: https://lovable.dev/• v0: https://v0.dev/• ChatPRD: https://www.chatprd.ai/• Figma: https://www.figma.com/• GitHub Copilot: https://github.com/features/copilot—Other references:• Jest: https://jestjs.io/• Vitest: https://vitest.dev/• MCP: https://www.anthropic.com/news/model-context-protocol• Confluence: https://www.atlassian.com/software/confluence—Production and marketing by https://penname.co/. For inquiries about sponsoring the podcast, email [email protected].

  --------  

  44:56

  --------

  44:56

More Technology podcasts

Trending Technology podcasts

About How I AI