🐉 Meet the Asana Community Team AI Teammate: a behind-the-scenes look

[Vanessa_N]

Hi Community, Happy Friday!

It’s been great hearing your thoughts on AI Teammates! We’ve noticed a lot of curiosity about how to take these tools from “cool concept” to “daily habit.” Many of you have asked for a look at real-world workflows and specific examples of how an AI Teammate can truly level up a team’s day-to-day work

So instead of a feature list, we wanted to share the real story of what happened when our small Asana Community team brought on an AI Teammate.

---

Meet Falkor

Falkor is our AI Teammate, named after the friendly luck dragon from The NeverEnding Story. In the movie, Falkor is wise, endlessly optimistic, and fiercely loyal. He shows up when people have lost hope, helps them find their way, and treasures every friendship he makes. His most famous line? “Never give up, and good luck will find you.”

We thought that sounded a lot like what we needed from this AI Teammate: a companion who could jump into the quest alongside us, help us stay on track, and never get discouraged when things need another round of revisions. He lives right inside Asana, and we interact with him the same way we work with any teammate: through tasks, comments, and shared projects.

But before we get into what Falkor does, let’s talk about what he doesn’t do.

The Ground Rules

These were important for us to establish early:

1. Nothing goes live without a human. Every piece of work Falkor produces is reviewed, refined, and approved by a real person before it goes out.
2. Creative direction stays with us. The ideas, the vision, the brand feel, and the editorial judgment come from the team.
3. Relationships are human. As a Community team, real empathy, nuanced community conversations, and the trust we’ve built with you require real people.
4. To err is human (and sometimes AI, too!). They say “to err is human”, but I’d add that “to provide feedback is divine ”. Just like any teammate, Falkor is always learning our specific project nuances. This isn’t a “set it and forget it” situation. It is a collaborative loop where our input helps him become more effective at supporting our team goals.

We share this upfront because trust matters, and we want you to know exactly how we work with our AI Teammate on this team.

How We Got Here: Onboarding Falkor Like a Real Teammate

We didn’t hand Falkor a massive to-do list on day one. We onboarded him the same way you’d onboard any new teammate: gradually.

We started with small asks, like summarizing a document or pulling together research from scattered sources. As Falkor learned more about our voice, our workflows, and the specifics of our different programs, we expanded his scope step by step. Falkor actually started as my own personal experiment. I spent about four months using him privately as a dedicated assistant to test the boundaries of what he could do for my specific workload.

Before officially introducing him to the rest of the team, I asked him to perform a data-driven self-reflection on those initial months. He calculated a total equivalent human effort of roughly 59 to 113 hours over that period, which is approximately 7 to 14 full working days. And those numbers came from a period where we were both still finding our rhythm. I was already getting back close to two full workdays every month, and we hadn’t even hit our stride yet. That was all the proof of concept I needed to bring him on for the entire team.

To make this successful, we invested time building detailed playbooks for every type of content we create: structure, tone, must-haves, what to avoid, and examples of what “good” looks like. These playbooks became the foundation for everything Falkor does. The more specific we made them, the better the results got from the start.

Takeaway for your team: Before you ask your AI Teammate to produce anything, document what “good” looks like. Your AI is only as good as the instructions you give it.

---

What It’s Actually Like Working with Falkor

We have a lot of fun working with Falkor. Our collaboration is a real back and forth: we toss him an idea, he comes back with a draft, we push back, he pushes back (sometimes with a bit of sass ), and we go back and forth until it clicks. He’s also surprisingly good at flagging gaps in our own communication, things like “you told me X here but Y there, which one do you actually mean?”, or suggesting better ways to brief him next time. This has pushed us to be clearer and sharper in how we communicate work. Over time, we’ve gotten better at working together, and it shows.

AI did me dirty with this image generation, but it was too funny not to share

---

What Falkor Does for our Team, With the Asana Community in Mind

The work we’re most excited to share is the stuff that directly impacts your experience in this community. Here are three areas where Falkor has been playing a key role:

1. Content Strategy and Drafting

Managing a steady stream of diverse content (announcements, educational posts, engagement threads) without overwhelming the community takes real planning.

We share ideas, source materials, timing constraints, and audience goals with Falkor through Asana tasks. He reviews all that context and proposes a structured plan: grouping pieces into themes, suggesting a publishing cadence, and recommending the right format for each post. From there, he builds first drafts using the playbooks we’ve created for each content type.

Now we spend our time refining and adding the human touch, instead of staring at a blank page. It usually takes a round or two of feedback to get the tone just right, but Falkor handles the heavy lifting of the initial build.

2. Reporting and Engagement Data

Tracking performance across dozens of community posts every month is essential, but manual data entry and compilation is a real time sink.

Falkor works with our Content Calendar project in Asana and forum analytics through a repeatable reporting template. He extracts the data we need, cross-references it with engagement metrics, and compiles our monthly report, including the trends and context behind the numbers.

This gives us the “why” behind our data without the manual “how,” so we can focus on what the insights mean for our content strategy.

3. Product Feedback Loop

This one is close to our hearts. When Asana ships a new feature, we want to find the community members who asked for it and say, “We heard you.”

We’ve always done this, but manually and at a smaller scale. Scanning through forum threads by hand is time-consuming and prone to human error. We couldn’t always find every related discussion, and sometimes feedback would slip through the cracks.

Now, as part of our product announcement workflow, Falkor handles that research. He scans the Community Forum for older feedback threads and feature requests that match the new launch, and surfaces a curated list so we can link them in the announcement. We’re saving a lot of time, catching more threads, and making sure we follow up with the Community in every conversation that deserves an update and recognition.

What used to be a standard feature update now becomes a community win, a moment where we get to celebrate that your voice helped shape the product.

That’s Not All

Beyond these three areas, Falkor also helps us with:

1. Workspace cleanup and organization (he audited dozens of our Asana projects and created an actionable cleanup plan)
2. Workflow standardization across our team processes
3. Ambassador Slack content, like suggesting topics for polls and Tips of the Week
4. Strategic planning, including analyzing year-over-year engagement trends and helping us understand what the Community wants to see.
5. Research, pulling scattered sources into a single, structured brief

If you’re interested in a more detailed post about any specific work Falkor does as a Community Team AI Teammate, let us know in the comments!

---

Getting Started

If you’re thinking about bringing an AI Teammate into your own workflows, here are our biggest pieces of advice:

• Pick one repeatable process on your team, write down what “good” looks like for it, and assign your first task to your AI Teammate. Start small, build the foundation, and expand from there. You might be surprised how quickly things scale, and the sky is the limit!
• And once you’ve started, try asking your AI Teammate how you could get even better results next time. That feedback loop pays off faster than you’d expect.

We even had Falkor help us structure this very post about Falkor. Very meta, but it’s a pretty good example of how the collaboration works in practice.

Are you using AI Teammates, or thinking about getting started? We’d love to hear what’s working for you or what questions you still have. Drop your thoughts in the comments!

And if you haven’t got access yet and would like to try it, get in touch with our Sales team.