The Day I Collaborated on My Professor's Passion Project

Nine weeks at Princeton. That was what the NSF REU program gave me: nine weeks in a computational cognitive science lab, working alongside people who thought about how humans thought. I walked in expecting to learn research methods. I walked out understanding something bigger about how ideas actually happened.

The Lab

The Velez CoLab at Princeton studied how people coordinated, adapted, and made decisions in groups. My advisor's work sat at the intersection of computer science, psychology, and game theory. The core question was simple to say and hard to solve: when you were on a team, how did you adapt to new information without breaking coordination with everyone else?

It sounded abstract until it mapped onto real life. A basketball team ran a play, one player spotted an opening and improvised, and suddenly everyone else had to decide whether to follow or stick to the original plan. A startup pivoted based on new data while part of the team kept shipping against the old roadmap. Same tension, different arena: individual adaptation versus group coordination.

My job was to help build the tools that could measure that tension directly. I worked on EXOBOUND, a behavioral research platform built in JavaScript with Phaser.js and jsPsych. In practical terms, it was a game designed to track how people made coordination decisions under different conditions. I modified mechanics, debugged asset loading, fixed trial sequencing, and kept documentation clean enough that the next person could actually use it.

The daily rhythm was fast and technical. A small code change could alter participant behavior, and a tiny UX issue could muddy an entire dataset. That made the work feel high stakes in a good way. Every decision had to survive both engineering logic and experimental logic.

The Moment It Clicked

About five weeks in, my professor invited me into a part of the project that was genuinely hers. Not intern maintenance work. Not curriculum. Her passion project. She had a new idea for an experimental condition and asked for my input on how to make it real.

That was the pivot point. The dynamic shifted from "execute this" to "think with me." She was not asking for a cleaner implementation of her idea. She was asking me to bring a different lens and pressure-test the concept with her.

We came at the same problem from different defaults. She started from theory and measurement precision. I started from a builder's instinct: what would this feel like for a participant in real time, where would confusion show up, and where would interface friction contaminate the behavior we were trying to study?

Those questions came from my background outside pure research. I had built swim programs, fitness training systems, and live events where user behavior decided whether the system worked. In each of those worlds, people did not follow your design because it was elegant. They followed it because it made sense under pressure.

That perspective turned out to be useful in the lab. It helped us identify moments where a participant was not making a coordination decision, but just decoding unclear instructions. It helped us catch places where the interface accidentally nudged strategy. My professor saw that quickly, and she treated it as signal, not noise.

What Collaboration Actually Looks Like

Research collaboration was not two people nodding at each other. It was two people disagreeing productively and staying in the room long enough to improve the idea. She would propose a structure, I would flag a technical constraint or design risk, she would reframe, I would build, we would test, it would fail, and we would run it again.

Some days were clean and fast. Other days were frustrating, especially when a bug looked like behavior or behavior looked like a bug. But even the frustrating loops were useful because they forced us to separate what we hoped the experiment was doing from what it was actually doing.

The best outputs came from that back-and-forth. They did not feel like her idea versus mine. They felt like a third thing that only appeared because her theoretical depth met my practical instinct in repeated cycles. That was the first time I felt collaboration at a level deeper than task-sharing.

Corporate teamwork usually means dividing a checklist and merging files. This was different. Real collaboration meant allowing your favorite version of an idea to get reshaped by someone else's perspective without treating that as defeat. It required trust, ego control, and a commitment to getting it right instead of getting credit.

What I Took With Me

I left Princeton with more than a line on a resume. I left with a clearer answer to how I wanted to work long term. I did not want to build in isolation. I wanted to build in environments where different disciplines collided and generated ideas no single lane could produce.

The experience also confirmed something I had suspected for a while. My unconventional background was not a liability in technical spaces. The years spent coaching, building businesses, and producing events trained me to notice operational friction and human behavior patterns that purely academic paths often missed.

In the right room, that difference became leverage. It gave the team a second operating system for evaluating design decisions. Instead of only asking "is this theoretically valid?" we could also ask "does this hold up when an actual person uses it under time pressure?"

That changed my definition of a strong team. The best ideas did not come from the single smartest person in the room. They came from the most useful collision of perspectives, applied with discipline and respect. That lesson stayed with me far beyond the program.

My professor's willingness to treat me as a collaborator, not just an assistant, set a standard I now carry into every project. When I lead, I look for people who think differently and can challenge assumptions without derailing momentum. That is where the breakthroughs live.

Closing

The research was published. The internship ended. But the operating model I took from that summer stayed permanent: build with other people, across disciplines, and combine perspectives that do not obviously fit together on paper.

Sometimes the most valuable thing you learned in a lab had nothing to do with the final dataset. For me, it was learning that collaboration was not a soft skill around the work. It was the work.