Most companies talk about AI adoption. Sendbird, the API platform for in-app chat, took it a step further. CEO John Kim and his team didn't just tell employees to use AI; they built an internal AI skills marketplace and turned it into a competitive sport. The goal? Stop duplication, spark cross-functional learning, and breed an entire company of 'AI gods.'
Key Takeaways
- Sendbird built an internal AI skills marketplace, akin to an app store, where employees create and share reusable 'plugins' and individual 'skills' to solve problems.
- The marketplace aims to create a "co-evolution" environment, as Kim puts it, preventing teams from rebuilding the same AI solutions in silos.
- Adoption was driven by a dual strategy: top-down encouragement from executive leaders like Kim and his CTO, combined with organic, bottom-up peer learning.
- To gamify engagement, Sendbird implemented an AI token consumption leaderboard, publicly ranking employees from 'AI newbie' to 'AI god.'
The Method
Sendbird’s approach to becoming an AI-first company involved building an internal platform, the 'Automator,' that hosts a marketplace for AI skills. Think of it like an internal app store, but for AI-powered mini-applications or prompts. John Kim explains, “So here anyone can create a plug-in. Plug-in is a collection of skills or you can create and download individual skills as well.” This means if a sales team creates a powerful AI prompt to draft customer emails, or an engineering team builds a code-summarization skill, it doesn't stay locked in their corner. It becomes a shared asset.
The core driver behind this system is efficiency and collective growth. Kim states their intention clearly: “we're trying to create this place where we can co-evolve rather than people operating in silos.” Instead of twenty different departments trying to solve similar problems with AI from scratch, they can leverage existing, proven solutions. This prevents the wasteful duplication of effort common in larger organizations as they rush to integrate AI.
Driving adoption wasn't solely a top-down mandate. While executives, including Kim himself and his CTO, actively championed the marketplace—“there's a both top down and bottom up. Top down meaning, you know, myself, a CTO, some of our executive leaders really tried to get people to adopt it”—the real traction came from the ground up. Teams that saw success with specific AI skills showcased their results, sparking curiosity and a desire for peer-to-peer learning. Kim notes, “There usually some skills involved. So I think there's that organic kind of pure learning aspect as well.” This organic spread, combined with a public AI token consumption leaderboard that ranks employees from 'AI newbie' to 'AI god,' created a self-reinforcing loop of creation, sharing, and competition.
Where This Breaks Down
This method shines brightest in companies with a strong engineering culture and existing internal tooling. A robust marketplace like Sendbird's needs a foundation, whether that's an internal platform team or readily available low-code AI tools. For smaller startups or organizations without dedicated tech resources, the overhead of building and maintaining such a system could outweigh the benefits.
The gamification aspect, while powerful, also carries a risk. If the leaderboard focuses purely on consumption or creation volume without quality checks, it could lead to a proliferation of low-value, poorly designed, or even redundant skills. It also relies on a culture of sharing, which might clash with highly competitive internal environments where individuals or teams prefer to guard their intellectual property. Without clear executive endorsement and a strong initial push, such a system could simply fail to gain momentum.
What to Do With This
Don't wait for a dedicated platform. Start simple this week: create a shared document or a private GitHub repository titled "AI Skills & Prompts." Task each team lead with contributing at least one high-value AI prompt or script their team uses. Then, pick a team to be the first "AI god" by publicly celebrating their most impactful shared skill in your next all-hands meeting.