Jarvis is real.

A few months ago I had the opportunity to speak at an AI Innovation Council event here in Charlotte, where I vide coded a functioning app (from scratch) in front of a live audience. While I was doing this on stage, my friend Charles D’Andrea (Managing Partner at Pattern50) was in the crowd. Without me knowing, he spun up a couple AI agents to work on the same goals. At the end, he walked up and showed me how his agents finished with a similar output in roughly the same amount of time. Wild! 😅

This little story is a great representation of the broader shifts that are happening across the industry. The ground is shifting from apps to AI agents. You can share your goals (in natural language) with an agent and it will choose the right tools and do the work on your behalf. It’s starting to feel more and more like Jarvis from Ironman actually exists (any MCU fans out there?).

As things continue to accelerate, interfaces will begin to learn/infer your intent and adapt in realtime AND our voices will become the mouse and the keyboard.

I firmly believe that the teams that will win in this next era will spend a lot of energy figuring out how to apply this new mental model and these capabilities to solve problems for their users/teammates/customers in new and novel ways.

Charles and I unpack this topic and more in the 1st episode of season 2 of the Building Great Experiences podcast. Catch the full episode and key takeaways from the conversation below.

*Quick heads up.. read to the end to check out everything we launched this past week with the StealthX 2.0 brand refresh 🤘

Agents are becoming the user.

In this episode, Charles and I explored how the interface is shifting from screens and menus to agents that can use tools like a teammate. As the capabilities progress, these agents will choose the right app for the job, connect to the data they need, and execute tasks end-to-end for you. Increasingly Agents will use software for us vs you actually using UI.

This may seem crazy, but it’s already happening today with general purpose agents that can browse the web, login to your accounts, fill out forms, and complete multi step tasks without custom integrations or special APIs (ex. GPT Agent and General Agents).

Also, this will become the primary way we interact with tools as voice-enabled tools like Siri, Alexa, and Google get more intelligent and humanity gets more comfortable talking out loud in front of others (and not feeling awkward doing it).

What this means is we’re quickly approaching a world where the experiences we design and build may not even be used by people, but will be used by someone’s AI agent on their behalf. This introduces a completely new paradigm that we’ll have to consider as we think through how to drive customer acquisition, engagement, and retention.

Putting this into action:

I know a lot of folks feel overwhelmed by the words “AI Agent.” It can feel like something only relegated to technical folks who know how to code. Not the case. Here are a couple ways to get started:

• List 5 tasks you do every week that are repeatable/recurring. Start by writing them down as simple commands a teammate could understand, then decompose them into a step-by-step process. For example, “Schedule a recurring invoice with a customer” or “Analyze this transcript, extract actions I committed to, and add them to [Project Management Tool].”

• Block 20-minutes every week to build a tiny agent-powered workflow using Zapier.com, Lindy.ai, Make.com, n8n.io, or Gumloop.com. Most of these tools now offer the ability to describe the agent you want to build, and it’ll figure it out for you, which is soooo much easier! I challenge you to ship something in that 20-minute block, even if it breaks. Learn by doing. No excuses 😉 

Personalization will shift from knowing the actions you’ve taken to understanding what you’re actually trying to do.

Charles and I talked about the move from behavior-based personalization to intent-based intelligent experiences.

For example, instead of simply knowing that a customer visited a sequence of pages on your website and you trying to make educated guesses and assumptions as to why they were there, AI will start to infer things like “Drew is planning a camping trip this weekend and it seems like he needs some suggestions from the products we offer.” Then it will be able to offer really valuable recommendations proactively vs the customer having to search/sort/filter and find on their own.

I believe that the best experiences in the future will interpret and infer someone’s needs and adjust proactively in real time.

Voice & vision will become everyday inputs.

Charles and I swapped stories on how we’re using a variety of tools daily that automatically capture/share meeting notes from recordings, turn messy thoughts during a commute into clear emails/docs/notes, and a lot more.

Charles shared an active client experiment where they’re using Gemini Live to scan rooms, identify maintenance items, and generate a plan a homeowner can follow. The path ahead is becoming increasingly more obvious.. talking and showing will beat clicking and typing, and we need to be thinking now about how we can be ready for this shift.

Also, wearables like glasses and pendants are increasingly getting lighter, smarter, and cheaper. I recently bought a Limitless pendant, it’s amazing. It captures conversations discretely and provides notes and feedback automatically. Several friends have the Meta Rayban glasses and love them. They record the world from their perspective and provide intelligent outputs.

There are obviously still privacy/security questions we need to sort out, but these types of tools are increasingly creating new ways to remove friction and elevate the experience, which we’ll all need to consider as we build great experiences in the years to come.

Putting this into action:

• If you’re not already doing it, pilot a meeting recording tool with your team. At this point in 2025, this is table stakes. There are a ton to choose from, just pick one and go. We use Fathom at StealthX but there are a ton of others out there that are just as good including Otter.ai and Fireflies.ai.

• Once you have a voice recorder setup for your team, connect it your knowledge hub (e.g., Notion, Google Drive, OneDrive, etc.), CRM (e.g., HubSpot), and project management tool/s (e.g., Asana, Jira, Notion, Monday, Clickup, etc.) so that your meeting notes and actions are automatically stored and accessible to the team.

Learn how to use AI… by asking AI how to use it.

If you don’t know how to use AI, ask AI to explain it to you. Use it as a tutor/collaborator. Ask it how to use it more effectively. It may sound simple/obvious now, but is often overlooked and is a huge unlock.

Charles and I both shared our experiences over the last few years using AI tools to learn AI (and other things) and how much it has accelerated our learning.

Putting this into action:

• Challenge yourself and everyone on your team to ask AI to teach them a new skill. For example, ask ChatGPT to do deep research on how to best to build a custom Agent (or use one that’s already built) that solves something you do every week. Ask it to break it down into a step-by-step process and teach you how to build/configure it.

• When you or a teammate learn something with AI, capture what you learned and share with others. Teach someone else to solidify the learning in your own mind.

• Check out NotebookLM. Create a notebook with content/context (e.g., eBooks, meeting notes, or whatever else you want to learn from) and have it create a series of podcast episodes to help you learn.

Make time & space weekly to practice as a team.

Charles and I both shared how we see teams getting overwhelmed by the crazy amount of new AI things coming out daily. The fix to feeling overwhelmed is by simply establishing a cadence for team learning, exploration, and experimentation. An easy way to do this is by meeting weekly for a show-and-tell and building one small thing together.

A favorite example I often share is team status updates. If your team is already tracking work in a project management tool, why not create a basic AI agent that assembles the update by itself? This is super easy to setup and saves a lot of manual effort to compile by hand.

Putting this into action:

• Hold a 30-60 minute session with your team every week to build together. Pick a small workflow and finish it in the meeting. Build small Agents and automations every week to get quick wins that compound over time.

• Use a simple nomination rule. If you copy and paste something twice in a week, log it as a candidate for AI/automation.

• Keep a visible scoreboard with your team to show progress. Track things like minutes saved, errors removed, time to resolution, etc. Have at least 1 simple metric to track and show value per agent/automation.

Wrapping up

There are a lot of exciting things on the horizon as AI continues to accelerate teamwork and enhance customer experience. As you consider the future, here are some questions to ponder:

• If an agent used your product today, what tasks could it complete end to end without a human, and where would it get stuck? What can you do now to get ready for this?

• What intent signals do you already have? How might you adapt your customer journey in real time based on those signals?

• How might voice or vision replace clicks and typing in your product or website? How can you start preparing for this paradigm shift now?

• What is 1 agent you can build today to save yourself or your team at least 1-2 hours per week?

Lastly, I have a simple challenge for you…

Ship 1 basic AI agent this week. It doesn’t need to be crazy. Just pick something simple/repeatable and do it. If you have a question about how to do this, just ping me. I don’t bite 😊 

Onward & upward,
Drew

P.s. If we haven’t met yet, hello! I’m Drew Burdick, Founder and Managing Partner at StealthX. We work with brands to design & build great customer experiences that win. I share ideas weekly through this newsletter & over on the Building Great Experiences podcast. Have a question? Feel free to contact us, I’d love to hear from you.

StealthX 2.0 is here! 🎉🎉🎉

After a lot of hard work, we shipped StealthX 2.0 last week that included a bunch of new goodness. In case you missed it, we:

• Sharpened our positioning and messaging to better align with how we work and talk.

• Designed, built, and launched a differentiated web experience with refreshed visuals.

• Released a new brand animation that showcases our new look and message.

• Launched season 2 of the Building Great Experiences Podcast with a new logo, intro/outro, and thumbnails.

• Unveiled our newest client story featuring our work to elevate the patient experience with our friends at OrthoCarolina.

Check everything out below: