Case Study: Building a Program That Changed How Faculty Taught

The problem

In the aftermath of the notable events of 2020, universities nationwide scrambled to respond to renewed attention on racial inequity. Almost universally, that response took the same shape: unconscious-bias training and interpersonal-tolerance workshops focused on individual attitudes and behavior. What was missing everywhere was a program that addressed something different and, in engineering and technology fields specifically, more consequential — how systems, algorithms, and processes can end up inaccurate or ineffective at solving the problems they’re built for, not just inequitable, when they’re designed without understanding the structural realities they operate within.

This is a textbook example of what’s known as a “wicked problem” — one with no clear boundaries, no agreed-upon definition, and no template solution, where even naming the problem correctly is most of the battle. That made it easy for the topic to be widely misunderstood and easy to avoid entirely, which is exactly why it had gone unaddressed everywhere else.

The reframe

I made the case that understanding historical and structural racism isn’t a soft-skills add-on to innovation work — it’s foundational to getting the work right at all. As I put it at the time: “trying to tackle any innovation or societal issue without understanding the impact of historical and structural racism is like trying to fathom the universe without knowing about carbon.” Carbon isn’t a minor ingredient — remove it from the picture and you get a fundamentally wrong understanding of how stars work, how old a fossil actually is, how photosynthesis functions, what living things are made of. The same is true here: build an innovation without accounting for structural realities, and the risk isn’t just an unfair outcome — it’s a solution built on an incomplete model of how the world actually works, and one that may fail at the very thing it was meant to do.

A well-known example: a medical triage algorithm built with no racial variable at all still consistently prioritized less-sick white patients over sicker Black patients, because the factors it used — like frequency of past doctor visits — were themselves shaped by structural disparities in access to insurance, flexible work, and child care. Nobody designing that algorithm was being biased. But the algorithm was also, quite simply, wrong at its actual job of identifying who needed care most — because the variables it used as proxies for health weren’t neutral — they carried the structural disparities baked into who gets to see a doctor in the first place.

That distinction — structural versus interpersonal — is what no other program addressed. We searched and found nothing comparable at any other university in the country.

The solution

I identified this need and built the program entirely on my own initiative — no one asked for it, and there was no existing mandate to fill it. Doing so meant navigating a genuinely sensitive topic while also selling the idea itself: to faculty chairs, faculty, the Dean of the School of Engineering, the Provost’s Office, postdocs, and staff across the university. Getting a program like this funded and adopted required as much stakeholder alignment and persuasion as it did subject-matter thinking.

One of the harder parts of that persuasion wasn’t the funding ask — it was framing. The moment people heard the topic, many assumed this was another round of the unconscious-bias and interpersonal-tolerance training that had already saturated higher ed for decades — what I came to think of as “their father’s DEI program.” It wasn’t, and the distinction mattered: this program had nothing to do with individual attitudes or behavior, and everything to do with how systems and technologies are actually built. Getting stakeholders past that first assumption, so they could engage with what the program actually was, took as much work as designing the program itself.

I conceived and secured $174K in funding for a first-of-its-kind program engaging Princeton faculty, postdocs, and staff in examining how historical and structural inequity gets embedded in the systems and technologies they build and study — run over three years and reaching 75 participants.

The outcome

• Multiple faculty revised their curricula, and one created an entirely new course inspired by the program
• Recognized with the Tiger Award, the highest honor for School of Engineering staff at Princeton
• Led to conversations with the Provost’s Office about scaling the program campus-wide

What made this work

Same instinct as the other case studies here: most institutions were solving the version of the problem that was easiest to see — individual behavior — because it’s the version everyone already had a template for. The harder, more useful question was further upstream: not “how do people treat each other,” but “does this actually work, given the world it’s being built into.” Naming a wicked problem clearly enough that people could actually act on it — rather than avoid it or reduce it to something smaller and more familiar — was most of the work. That’s the question this program was built to answer — not just for the sake of fairness, but for the sake of getting innovation right.