Leading With AI: What Helped Me Through the Hard Parts
Real moments, practical lessons, and quiet wins from leading the Monocle.com rebuild.
The new Monocle.com site recently launched — the result of a long journey filled with highs and lows, frustrations and wins, across every discipline. It stands out as one of the most challenging projects I’ve worked on at 10up — not just because of its technical complexity and integrations, but also in how we worked together as a team. The process, the resourcing, and especially the surge in conversation around AI made it a unique experience.
I’ve always been fairly pro-AI. I don’t see it as a threat to my role — I see it as a tireless teammate that helps me work smarter, move faster, and be more effective. In this post, I want to share how I used AI throughout the Monocle project — not just in engineering, but in areas that go beyond code.
Here are five ways I leaned on AI to bridge gaps, accelerate progress, and stay steady under pressure — with real examples to show that, when used thoughtfully, AI really can make us better at what we do.
Gut-Checking Technical Decisions
One of the hardest parts of leading a team is making decisions in the face of ambiguity — and doing it quickly. I used AI as a sounding board to validate my thinking when navigating unfamiliar patterns, edge cases, or murky technical decisions. It gave me a “second brain” to bounce ideas off, which helped reduce decision fatigue and let me move forward with more confidence.
It wasn’t just about getting the right answer — it was about reducing the emotional weight of needing to constantly interrupt others for validation. We’ve all had those moments where you want to ask a Web Engineer, “Hey, does this make sense?” but you hold back — not wanting to slow them down or feel like you’re adding pressure. AI helped me bridge that gap. It offered an initial gut check, so that by the time I did ask a teammate for input, I’d already sharpened my thinking and felt more prepared for the conversation.
Experience:
During the Monocle build, we integrated a relatively new headless eCommerce platform called Swell. It needed to communicate with Piano to handle user authentication — and on the frontend, we were layering Alpine.js for reactivity. I had a solid grasp of Alpine’s model, but the orchestration between Swell and Piano wasn’t immediately clear.
Before jumping into a call or writing up a bunch of questions, I used AI to gather more context on both systems — their tech stacks, how their APIs worked, even the general approach their teams seemed to favor. Then I prompted AI to generate a minimal but robust example of how the two platforms could integrate. That context helped me mentally map the interaction and move forward with clarity.
Rapid MVPs Without Burnout
On high-pressure projects like Monocle, you don’t always have the luxury of time — especially when spinning up new features to help drive decisions or unblock other teams.
Rather than burn hours on something that might change completely after one stakeholder review, I started using AI to help scaffold MVPs. I’d describe what I needed — a “Saved Views” panel, a reactive dropdown filter, a basic TypeScript schema — and it would return a clean starting point I could tweak and wire up quickly.
This wasn’t about cutting corners. It was about preserving energy for the parts that required deeper thinking or polish — while keeping the project moving forward.
Experience:
The team requested a custom ad unit inspired by Wimbledon’s Rolex sponsorship — specifically, a live “ticking” clock face synced to the user’s local timezone. After inspecting the Rolex example, I realized how complex this could be: real-time sync, time zone logic, animation performance, accessibility… the list went on.
With AI, I quickly broke the feature down, explored animation strategies, and generated a working MVP. It helped us estimate more accurately and kept pressure off me — while still delivering something high-impact.
Faster Debugging with AI
We’ve all had those bugs — subtle, persistent, and completely undocumented. The kind where you open ten tabs, hoping one StackOverflow post from 2013 holds the key.
AI changed how I approached debugging. I could describe the issue, paste a snippet, and quickly get back pointed analysis or even an overlooked edge case. It saved time, but more importantly, it helped me understand why things were breaking faster — and that’s what really counts.
Experience:
Monocle requested a horizontally scrollable component — a pattern I’ve grown to dread on the web. Unlike mobile apps, the browser ecosystem is full of quirks that break even the cleanest sliders.
I built a CSS-only solution enhanced with JS for accessibility. It worked well — until bugs popped up on Safari. Rather than dive down the rabbit hole, I asked AI to help diagnose it. Within minutes, it pointed me toward a known 100vw issue with Safari rendering. That call saved me hours of trial and error.
Stepping into Backend Territory
Frontend engineers usually translate data into UI. But on Monocle, that line blurred. With backend engineers stretched thin, I often stepped in — writing PHP, digging into unfamiliar parts of the theme, and doing my best to help without slowing anyone down.
AI became a kind of guide — helping me understand Fueled’s backend conventions, structure logic correctly, and write clean code that wouldn’t need a ton of rewriting.
Experience:
There were a few weeks where I stepped up, not as a Backend Engineer, but as a Frontend Engineer willing to help. I’ve worked with PHP and WordPress APIs for years, but Fueled’s PHP standards keep evolving — every project more abstracted, more refined.
I wanted to contribute meaningfully. So I asked AI to walk me through Fueled’s best practices and assist with a small feature I was building alongside Konstantinos. I was nervous about the PR — but it passed without a hitch. It was a small win, but a powerful reminder that with the right support, we can stretch into unfamiliar territory and still deliver quality work.
Clearer Communication Under Pressure
Leading through a long, demanding project carries emotional weight. And sometimes, that stress creeps into how we show up in Slack, code reviews, or project updates. I didn’t want that — not for myself, and not for the team.
So I started using AI to gut-check my messages. I’d write them raw, then ask AI to help me refine them — make them calmer, clearer, and less reactive. Not fake. Just… clearer. It became a small but powerful practice.
Experience:
This might be a contentious point, but relationships are complex. Communication is hard. And when the pressure’s high — the team is tired, clients are shifting goalposts — you sometimes just need a little help.
It wasn’t support I lacked (shoutout to Gabe, Lily, and Brent) — it was clarity. AI helped me take a breath, tighten my language, and avoid letting frustration seep into the way I communicated. I tend to ramble when I’m stressed. AI helped me strip away the noise and get to the point. I even learned a few new ways to say difficult things… more kindly.
Final Thoughts
Looking back, I didn’t use AI to replace my skills — I used it to reinforce them. It helped me move faster when time was tight, get unstuck when problems were murky, and communicate more clearly when the pressure was on.
What surprised me most wasn’t the technical insight — it was the quiet confidence it gave me. Whether it was gut-checking a feature, scaffolding a tough integration, or rewriting a message, AI became a silent partner in the background — helping me lead more effectively.
This wasn’t about shortcuts. It was about showing up better — and getting the job done well.
