How I AI: John Kim's Playbook for AI Transformation with Quests, Skills, and 'AI Gods'
Delight.ai CEO John Kim shares his revolutionary approach to company-wide AI adoption, featuring an internal automation marketplace, a skills library, and a gamified token-tracking dashboard that gets everyone building.
Claire Vo
In this episode, I was so excited to sit down with John Kim, the co-founder and CEO of Delight.ai. I often tell leaders that to truly transform their company, they need to think of their team as a product. John is one of the few people I've met who has actually put this into practice, turning AI adoption from a top-down mandate into an internal product that's fun, engaging, and incredibly effective.
What John and his team are building is a blueprint for any organization that wants to move beyond simply telling employees to 'use AI' and instead empower them to become builders. It's not about doing more with less; it’s about having a bigger ambition, building more creative and delightful experiences, and fundamentally changing how work gets done. John’s approach shows that when you give people the right tools, guardrails, and incentives, incredible things happen.
We explored the internal platforms he’s built to foster this culture, from a gamified 'Quest' system for requesting automations to a dashboard that tracks every employee's journey from 'Beginner' to 'AI God.' It's a pragmatic and inspiring look at what it takes to build a true AI-first company from the ground up.
Workflow 1: Building 'Automators', an Internal AI Marketplace
The biggest challenge for AI adoption isn't a lack of ideas; it's the friction between an idea and its execution. To solve this, John's team built the 'Automators Platform,' an internal marketplace designed to connect AI needs with AI builders.
It’s a brilliant system that sidesteps the traditional, often cumbersome, engineering prioritization process. Instead of ideas dying in a Jira backlog, they become 'Quests' in a gamified system anyone can participate in.
How the Quest System Works
The process is straightforward and empowering:
- Create a Quest: Any employee, from finance to sales, can raise their hand and create a 'Quest.' This is a request for an automation or tool. For example, the finance department could request a workflow to automate accounts receivable.
- Define the Need: The 'Quest Giver' specifies what they need, providing the subject matter expertise. They know the problem intimately, even if they don't know the technical solution.
- Form a Party: Other employees—engineers with downtime, or AI-enabled marketers—can see the open quests and offer to help. It creates a dynamic where people can team up to tackle problems they find interesting.
- Build and Submit: The team collaborates to build the solution. The result is usually a link to a GitHub repository or a new internal skill.
What makes this so powerful is the transparency and the focus on value. Each quest publicly displays its estimated business impact.
'You're kind of like building this shadow AI roadmap that works really efficiently. Basically a marketplace of AI needs and AI builders inside your company where anybody can just pop in and say, oh, I, I think I know how to do that.'
Gamification and Rewards
To fuel engagement, the Automators platform is gamified.
- Experience Points (XP): Completing quests earns participants XP.
- Rewards: This XP isn't just for bragging rights. It can be exchanged for real-world rewards like gift cards, a tea meeting with any executive, or the chance to present their work at the weekly company-wide standup.
This system celebrates builders from all corners of the company. As John noted, the people presenting their creations are often from marketing, recruiting, or sales—not just engineering.
Workflow 2: Letting Marketers Cook with Secure App Templates
This internal system has led to some amazing projects. The most striking example is a fully functional swag store, 'Big S Energy,' built entirely by the marketing team in just a couple of days with no engineering support.
The store sells witty, culture-focused merchandise, integrates with Stripe for payments, and even includes a secret Easter egg for conference details, unlocked using the classic Konami Code.
In a traditional company, an idea like this would likely be deprioritized or relegated to a clunky, out-of-the-box solution. By empowering the marketing team, they were able to ship something fun, creative, and delightful for customers. It's a perfect illustration of a point made by a previous guest, Jason Levin: 'Let your marketers cook.'
The Happy Path to Production
How can a non-technical team ship a production-ready e-commerce site? The key is providing a secure, templated 'happy path.' John's team created internal app templates where all the complex and sensitive components are pre-configured.
- Authentication & Security: Already built-in and vetted by InfoSec.
- Environments & Databases: Pre-configured so builders don't have to worry about infrastructure.
- Compliance: All security and logging requirements are handled by the template.
A marketer or salesperson can just grab a template, build their idea on top of it, and push it to production safely. This removes the single biggest barrier for non-technical builders: the fear of breaking something or creating a security risk.
Workflow 3: Measuring What Matters: The AI Usage Dashboard and 'AI God' Tiers
To drive this transformation, you have to measure it. John showed me their internal AI usage dashboard, which provides a comprehensive view of AI adoption across the company. This isn't about punishment; it's about understanding the journey and providing targeted help.
From Beginner to 'AI God'
The dashboard gamifies usage through a five-tier system based on daily token consumption. Every manager knows where their team members stand and can provide the right kind of enablement.
- Beginner
- Intermediate
- Expert
- Architect/Catalyst
- AI God (over 100 million tokens a day)
This framework makes AI fluency an aspirational goal. It's okay to be a beginner, but the organization provides a clear path to becoming an expert. The top token consumers at the company are the executives, including the CTO and John himself, signaling to the entire organization that this is the new standard.
Smoothing the Curve
One of the most forward-thinking metrics John tracks is the 'smoothness of the curve.' Dips in the token usage graph indicate weekends or vacations when humans aren't working. The ultimate goal is to smooth out that curve, which means AI agents and automations are working autonomously around the clock, filling the gaps.
Workflow 4: John's Personal AI Workflows for Learning and Knowledge
John doesn't just preach AI usage; he lives it. He shared a couple of his personal workflows that showcase the power of AI for individual productivity and learning.
The Gardener: An AI for Your Notes
John created and open-sourced a project he calls 'The Gardener.' It's an automated script for personal knowledge bases like Obsidian. Every day, the Gardener agent combs through his notes to:
- Fix typos and grammatical errors.
- Enrich notes by researching people and companies mentioned.
- Create headings, clusters, and cross-links between related ideas.
- Nurture 'seedling' notes into mature documents.
It’s like having a personal librarian who constantly organizes and improves your second brain.
Creating Personal Learning Centers
My favorite personal workflow John shared was how he uses AI to create entire personal learning centers on complex topics. He gave an example of creating a deep-dive course on neuroscience.
He uses a detailed prompt with Claude Code or Codex to generate a complete, structured, and interlinked knowledge base.
```
You're like a PhD, neuroscience researcher. Here's what you're trying to create...
```
Within 20 minutes, the AI returns a beautiful, explorable website covering everything from key neuroscientists to neurological disorders to neuromodulators like dopamine and serotonin. He's done this for neuroscience, quantum mechanics, and fusion energy.
This is such a powerful application of AI for personal growth. You can create a bespoke curriculum on any topic, tailored to your brain, and explore it in a novel way. It’s an opportunity to learn things we never could before.
Final Thoughts
John's work at Delight.ai is a masterclass in driving organizational change. The key lessons are clear: treat internal adoption as a product, empower your creative teams, build secure paths to production, measure what matters without shame, and lead from the front.
By creating a system that is both fun and impactful, he has unlocked a new level of productivity and creativity. It's a reminder that this new era of technology is not just about tools, but about the culture and systems we build to support them. It's a future where everyone has the power to be a builder.
Thank you to our sponsors
- WorkOS—Make your app Enterprise Ready today
- Thoughtspot—Build AI-powered analytics into your product
Episode Links
- YouTube: https://youtu.be/uH39OZ-KnkY
- Spotify: https://open.spotify.com/episode/3E7OmI3CD1bhuqVCg7JNvA?si=reDhjhAvSQKv5-p9ELUvIw
- Apple Podcasts: N/A
