Showing customers the art of what’s possible: A conversation with an AI Deployment Strategist

AI deployments don’t succeed on technology alone — they succeed when people, workflows, and systems actually change together.

In this Q&A, Bea, an AI Deployment Strategist at Assembled, shares how she moved from customer success into a hands-on AI role, what it takes to deploy AI products with real customers, and how change management, technical curiosity, and cross-functional collaboration all factor into making AI work in practice — not just in theory.

This conversation has been edited for brevity and clarity.

Q: How did you become an AI Deployment Strategist at Assembled?

I started as a Customer Success Associate supporting major customers like DoorDash and Salesforce. The transition happened pretty organically through conversations with my manager about what parts of the work I was really enjoying.

Coming from management consulting, I knew I wanted to work at a smaller tech startup where I could explore different areas. The CSA role was intentionally broad, giving me exposure to both customer success and implementations, with the idea that I'd eventually gravitate toward whatever clicked.

The AI opportunity came up when the team needed someone who spoke Spanish to help with a customer deployment. What started as 25% of my time on that project became 100% within three weeks. I loved how hands-on it felt — building workflows, collaborating closely with engineering, and really shaping the product based on customer needs.

When the Deployment Strategist role opened up in July, it was an easy decision. I was already doing the work and loving it.

Q: What does an AI Deployment Strategist actually do day-to-day?

At its core, it's similar to an implementation role. We work with customers to deploy our AI products — whether that's integrating with their CRM for our agent products or training their support agents on Copilot. We make sure they're getting the most out of the product and seeing real efficiencies and ROI.

But here's what makes it different: we're often working with teams embarking on their first AI project. Many don't fully understand what AI can do, or they've received a mandate from leadership to "automate more" without a clear roadmap.

A big part of my role is helping them understand the art of the possible — not just answering basic questions, but thinking creatively about how to integrate with their existing systems and truly transform their workflows.

These projects typically run two to four months and involve both tactical implementation work and that blue-sky thinking about what's possible. My goal? Help teams offload mundane, repetitive tasks to AI so agents can focus on more meaningful work — like actually connecting with customers instead of resetting passwords for the tenth time that day.

Q: Can you share an example of how AI transforms the agent experience?

Yesterday, I was talking with a music distributor that works with smaller artists on platforms like Spotify and Apple Music. Their support agents are mostly artists themselves, and they told me they wish they could spend more time connecting with other artists on a creative level instead of handling routine requests like password resets.

That's exactly what we're trying to enable.

It doesn't make sense for an agent to waste time switching between multiple screens doing lookups when we can automatically surface that information. Or to answer the same basic question ten times a day when AI can handle it.

I think about how I've used AI tools to eliminate mundane tasks in my own work, and I want the same for support agents. When we automate those repetitive tasks, we give agents back time to do what they do best — build real connections with customers.

Q: What role does change management play in AI deployments?

Change management is huge, and there are some interesting dynamics at play.

Sometimes the team leading AI efforts isn't the support team — it's an ops group — which can create tension. Support team managers are rightfully concerned about what AI means for their roles and their teams. They're thinking, "I don't want to fire my entire team."

There's also an interesting tension between our WFM and AI products. I recently had a customer say, "We want to automate X percent more with AI, but you realize that means we'll need fewer seats on the WFM side." It's a real give-and-take that requires careful navigation.

A big part of my role is shepherding these conversations within organizations, building connections between different teams, and helping them see the value on both sides. I work closely with the CSM on the WFM side to manage these "meetings of the minds" and figure out what a world with both WFM and AI looks like.

I also try to stay on top of everything happening in the AI landscape so I can be that expert resource. Most of these folks don't have time to track every new model that comes out or understand what it means for them.

Q: How do you get people excited about AI instead of feeling threatened by it?

I think the key is helping people see how they're already using AI in their own lives. At this point, almost everyone has used ChatGPT, Gemini, or another chatbot.

I position it in terms of the tasks AI currently excels at — and will continue to excel at for the next few years — which are those really basic, boring tasks. No one wants to spend their day doing huge Excel analyses or resetting passwords.

By automating or speeding up those rote, menial tasks, we unlock time for people to either go home early to see their kids, brush up on skills they're interested in, connect more meaningfully with customers, or take more time on complicated tickets they currently have to rush through

The reality is that every customer I've talked to in the past two months has said, "Our queue is insane right now and our whole team is underwater. How can we increase automation?"

Most teams are actually understaffed right now, so AI becomes a way to deliver relief and make their day-to-day easier, not a threat to their jobs.

Q: How do you manage the technical side of your role while juggling so many stakeholders?

It's definitely one of the more challenging aspects — balancing stakeholder needs with limited engineering support.

We work with stakeholders to prioritize what will deliver the most impact, then take that back to the engineering team with a clear business case for why something needs to happen.

As Deployment Strategists, we individually create tickets with detailed information, then we prioritize as a team before taking it to engineering. On Mondays, we review with all the engineering PMs to align on what can and can't get done.

What's been a huge unlock recently is using tools like Cursor to make updates myself. There are so many items that don't classify as critical but are still major pain points for customers.

Now I can make those updates and test them — I still need an engineer to sign off, but it takes 90% of the workload off their plate. It's incredibly satisfying to unblock customers on things that otherwise might never get done or would take five months.

Q: What's it like contributing code when you came from a non-technical background?

It's pretty surreal! I never imagined when I joined as a CSA that I'd be shipping code.

What I've appreciated most about Assembled is the opportunities to learn and the leadership team's willingness to not box you into a role. One of our principles is "Assembled is my team and title," and I initially thought that just meant doing support rotations or helping out where needed.

But it's so much more than that.

I've never been told no when I've wanted to take something on. Instead, I've been empowered to make changes. They gave me a new laptop when mine didn't have enough capacity for engineering testing and upgraded my Cursor plan so I could do more coding. It's always been "How can we help you?" never "That's outside your scope."

It's crazy to think that in just over a year, I've gone from management consulting to being an AI Deployment Strategist who ships code. That path would probably take 20 years at a Fortune 500 company.

People here are so busy, but they're incredibly generous with their time and excited to help you learn. There's really no concept of a strict role — it's more like, whatever you can do that benefits our goals, do it.

Q: What qualities make someone successful as an AI Deployment Strategist at Assembled?

Curiosity is the most important quality.

The product is still new, so you need a willingness to learn and recognize that not everything will be laid out for you. You have to be comfortable doing some digging and getting your hands a little dirty.

But if you're curious, that's actually a highlight of the role — it's what makes the job interesting. You're not just following a playbook; you're helping shape what's possible with AI for support teams.

That combination of technical work, customer interaction, and creative problem-solving requires someone who's genuinely excited to explore and learn.

Want to join the teams building and deploying AI at Assembled? Check out our open roles.