Amirreza Tootchi, a Phd candidate in Mechanical Engineering who is studying in Indianapolis.

Amidst all the hype about AI and machine learning, you also hear stories about the computational models “hallucinating” and getting things horribly wrong. As this new world rapidly unfolds, Amirreza Tootchi wants to bridge human trust and machine intelligence by quantifying machine learning model uncertainty in engineering design. The PhD candidate in Mechanical Engineering, studying in Indianapolis, is building a mathematical “shield” around this uncertainty to ensure that AI-designed products like bridges, robots, aircraft, and the like perform consistently and reliably.

What is your research focus, what investigative avenues are you pursuing?

I’m riding the wave between machine learning and mechanical design under the brilliant mentorship of Professor Xiaoping Du, a pioneer in uncertainty-based robust design.

But let’s not make it sound robotic.

I’m not just “pursuing a PhD.” I’m chasing a scientific puzzle that keeps me up at night; how can we trust machine learning models when they themselves are uncertain? My research zooms into this exact black box: quantifying machine learning model uncertainty in engineering design, where the stakes are real — bridges, robots, aircraft — not just lines of code.

Imagine this: a design engineer relies on a surrogate ML model to predict structural performance. It works great … until it doesn’t. My work builds a mathematical shield around this uncertainty—using statistical modeling, probabilistic analysis and robust optimization to ensure that products perform reliably even when the model gets shaky.

To get there, I combine everything from Gaussian process regression — a probabilistic method to estimate predictive value and uncertainty — to Taylor-based reliability theory, which approximates uncertainty to increase engineering reliability, along with advanced simulation frameworks. It’s like being a control freak for algorithms — I make sure their unpredictability doesn’t sink the ship.

The bigger picture? We’re building a future where AI-designed systems are not just smart but safe. Think autonomous vehicles that don’t crash in weird edge cases. Medical robots that don’t guess. Spacecrafts that survive chaos.

And if you ask me what I hope to achieve — here it is, plain and simple: I want to build a bridge — not between two cities, but between human trust and machine intelligence. Because in this AI-infused world, uncertainty is no longer a nuisance — it’s the frontier. And I’m here to map it.

What spurred your interest in this particular topic and line of research?

Funny enough, it all started with doubt.

Not the kind that makes you question yourself — the kind that makes you question the system. I’ve always had this inner voice that whispers: “But… what if it doesn’t work perfectly?” Whether it was a robot arm jerking mid-motion or a predictive model giving confident but wrong results, I found myself obsessed with what others brushed aside: uncertainty.

Back in grad school, while others were focused on crafting clean mathematical models of their robotic or mechanical systems, I was wrestling with a brutal truth: real-world models are messy. Uncertain. Incomplete. No matter how detailed your differential equations are, there’s always something off — noise, friction, nonlinear behavior, something unmeasured.

So I turned to intelligent control systems — fault-tolerant control, robust control, adaptive strategies — to handle the fact that my models weren’t perfect. That became the heart of my MSc thesis: designing controllers for pneumatic systems that could survive and thrive despite model uncertainties. The goal was clear: if I couldn’t fix the model, I’d build controllers that could handle its flaws.

But deep down, a question lingered: “What if, one day, I could actually make the model better?”

Fast forward to my PhD — and that’s exactly what I’m doing.

Now, instead of patching over uncertainty with clever control tricks, I’m attacking the source head-on. I’m exploring how machine learning models — powerful but imperfect — can be made trustworthy by quantifying their own prediction uncertainty. I’m flipping the script: not just living with flawed models, but actively improving how we understand, quantify and use them in design and decision-making.

Then I discovered robust design under uncertainty — and suddenly, it felt like the field had been waiting for me this whole time.

But the real turning point? It wasn’t a paper. It was a failure.

During my MSc thesis, I built a fault-tolerant control system for a pneumatic servo mechanism. It worked beautifully… until it didn’t. One minor unmodeled disturbance, and the whole thing wobbled like a jellyfish in a blender. That’s when it hit me: engineering isn’t just about building strong systems — it’s about making them fail gracefully.

Today, in my PhD, I work at the intersection of machine learning and reliability engineering, helping systems prepare for the unexpected. Because let’s be honest, ML is amazing, but it's also like a really talented intern: helpful, fast and occasionally catastrophic.

So why this topic? Because in a world increasingly ruled by algorithms, I chose to be the one asking: “What if it’s wrong?” And then designing for that very moment.

Why did you choose Purdue to continue your studies as a graduate student?

Why Purdue?

Let’s be honest, every ambitious engineer knows that Purdue isn’t just a name. It’s a gravitational field. A place where legendary meets possible. But for me, the choice went far beyond rankings and reputations.

At the time I was offered a faculty-track position back home — imagine that: job security, respect, the whole academic package. But I wanted more than a title. I wanted transformation. I had this unshakable feeling that I hadn't yet become who I was meant to be — that I needed to leave comfort to enter challenge.

Purdue felt like the perfect paradox: rigorous and welcoming, grounded and cutting-edge. It had Professor Xiaoping Du, whose work in uncertainty quantification felt eerily aligned with the mental battles I’d been having during my MSc years — the "what ifs" that haunted every imperfect model I ever built.

But it wasn’t just about academics. Purdue had something I hadn’t found anywhere else: a culture of cross-pollination. Where else could a mechanical engineer dip into AI summits sponsored by Google, get nominated for an Apple Fellowship, and then walk into a lab meeting discussing biomechanics, robotics and brain-computer interface systems in the same breath? Purdue’s environment practically begged me to dream bigger — and then gave me the tools to make it real.

Then came the funding. I received a PhD Fellowship, which was more than financial support. It was a signal: We believe in you. That matters when you’re thousands of miles from home, rewriting your identity in a foreign country, trying to turn doubt into discovery.

And finally — Indianapolis. A city that surprises you. Affordable, full of innovation, diverse in its people and its possibilities. It didn’t distract — it focused me. Gave me breathing room to build, to research, to evolve.

So no, I didn’t just choose Purdue. I chose becoming someone I hadn’t met yet. Purdue said, “Let’s meet him together.”

When did you first get interested in engineering and science?

You could say it started with a whisper not a thunderclap.

I wasn’t the kid blowing up toasters or building robots from scrap metal. No. I was the quiet one in the back of the class, asking questions no one else thought to ask. Why does this fall? What’s behind that movement? What if you bend the rules of how things work?

By the time I reached 5th grade, I was already being called “the professor” by classmates — not because I had the answers, but because I couldn’t stop chasing the questions. That same year, I took the nationwide National Organization for Development of Exceptional Talents (NODET) entrance exam — a brutal, two-stage test in math, science and logic. I was the only student from my school to pass. That moment cracked open a new world.

NODET wasn’t just a school. It was Hogwarts for the mathematically obsessed. For seven years — middle and high school — I was surrounded by prodigies, taught by teachers who didn’t just lecture, they sparked revolutions. One of them, my physics teacher in high school, didn’t just teach Newton — he performed him. Suddenly, physics wasn’t a subject. It was a language. And I was fluent.

That’s when it clicked: If physics is how the universe thinks, then mechanical engineering is how humans respond.

It became my gateway to everything — robotics, neuroscience, AI, control systems — a single discipline that let me collaborate with every other. Mechanical engineering, to me, was never narrow. It was the center of the spiderweb.

Looking back, my interest in engineering wasn’t one moment. It was a series of quiet obsessions, each one layering on the next. And eventually, they whispered the loudest truth of all: You were born to build what others only imagine.

What’s it like being a Purdue graduate student in Indianapolis?

Imagine being dropped into a room where everyone speaks your language — not just technically, but emotionally. That’s what studying in Indianapolis feels like.

My advisor, Professor Xiaoping Du, doesn’t just guide research. He orchestrates insight. Working with him is like playing chess with someone who knows 10 moves ahead yet still lets you grow into your own strategy. He gives you space, but not the kind that lets you float aimlessly. The kind that dares you to rise.

Our department is filled with people who don’t just publish papers, they live them. You see it in the sparkle of their lectures, the pointed clarity of their questions, the subtle smirk when they know they’ve just challenged you a little too hard on purpose.

And the students? We’re not just lab partners. We’re co-conspirators in the science of problem-solving. I’ve had late-night debugging sessions turn into philosophical debates about the future of AI in design. I’ve seen casual lunch breaks evolve into spontaneous whiteboard battles — equations flying, jokes slipping in between Greek letters and neural nets. There’s an unspoken bond in the air: we're all here because we know this is where the work becomes art.

Indianapolis, too, plays a role. It’s not loud like New York or glossy like Silicon Valley, and that’s its secret weapon. It lets you focus. It gives you access to growing smart infrastructure initiatives, connections to performance and analytics companies, and a blend of affordability and opportunity that lets students build instead of just survive.

To me, Purdue isn’t just an academic institution — it’s a launchpad. A community where talent meets trust. A city where innovation whispers louder than noise. And a place where I’m building the most thrilling chapter of my life — one equation, one idea, one person at a time.

What else have you learned at Purdue, beyond deepening your knowledge of subject matter?

Oh yes — Purdue didn’t just teach me what to research. It taught me how to think in a way that makes research feel like jazz: structured, bold and always one beat from surprise.

Before Purdue, I thought research was linear — like following a GPS. Now I know it’s more like orienteering blindfolded in a jungle with a pencil and a dream.

I’ve learned that the scientific method isn’t a checklist — it’s a rhythm. You don’t just ask a question, form a hypothesis and go. You wrestle with the unknown. You live with doubt. You test something 10 times, and just when you're about to give up, the 11th try teaches you more than the first 10 combined.

One of my biggest realizations? Ideas are fragile. They show up whispering in the middle of data analysis or right before you fall asleep. So, I learned to catch them, document obsessively and treat every weird idea like a potential breakthrough because sometimes, it is.

I also learned that great research isn’t done alone. Collaboration at Purdue is not performative — it’s genuine synergy. I’ve worked with people who make you sharper without saying a word — just the way they approach problems changes how you approach yours. We don’t compete — we cross-pollinate.

And then there’s structure — the invisible art of managing chaos. Purdue taught me to create flow: how to break down complex tasks into beautiful, executable timelines; how to chase multiple research threads without unraveling; how to build systems of thinking that actually scale with your ambition.

So, yes, I’ve learned new equations and models. But more than that, I’ve learned how to lead an idea from spark to reality and how to take a failure, laugh at it, learn from it and let it fuel the next iteration. Because that’s the secret sauce of research: Not perfection. Process!

What is the Purdue research environment like?

Purdue’s research environment? Think NASA lab energy meets quiet Zen temple! There’s this beautiful tension — high-stakes innovation on one side, deep reflective thinking on the other. It’s a place that welcomes your craziest ideas and then dares you to prove them with math.

One of the first things I fell in love with here was the freedom to explore deeply, not just widely. No one here’s interested in surface-level success. We go down the rabbit hole and come back with results.

That’s exactly how my PhD project took shape: I started with a question about uncertainty in machine learning, and it grew into a published paper in the ASME Journal of Mechanical Design. That work wasn’t just technical — it was a statement. We proposed a method to protect engineering design from the unpredictability of AI models. The impact? Bringing safety and conservatism back into AI-assisted systems — something the world desperately needs.

In parallel, I’ve had the chance to teach and mentor students, and I say this without exaggeration: some of the most meaningful moments in my PhD haven’t been in front of a laptop but in front of a classroom. Whether it’s explaining robust design to undergrads or helping a student debug a controller in Simulink, a simulation software, at 9 p.m., that connection — from learner to learner — is powerful. I’ve taught courses like Control Systems, Statics, Machine Design, and Robust Design with ML, and each one taught me as much as I taught them.

I’ve also been privileged to be honored with awards. I was awarded the Competitive Fellowship for PhD students at Purdue. I was selected as a Purdue/Google AI Summit Fellow. I even received a nomination for the Apple Fellowship — which felt surreal. It was like the universe saying: “Yes, your weird, wonderful mix of mechanical engineering, AI, neuroscience and control theory actually matters.”

As for commercialization? My work touches the threshold. I’m not developing a product, but I’m laying the theoretical foundation for safer ML integration into design systems. The kind that future patents will likely cite.

To put it simply: Purdue gave me the tools. It gave me the stage. And then it whispered, “Go build something only you could build.”

What advice might you give to other students deciding where to attend graduate school?

First, a disclaimer: grad school isn’t Hogwarts. There’s no magical hat to sort you. No single professor who waves a wand and transforms your life.

But if you’re lucky — really lucky — you find a place that doesn’t just sharpen your skills. It sharpens your identity.

That’s what Purdue did for me.

So, here’s my advice to any student standing at the edge of the grad school cliff, wondering where to jump:

Don’t chase prestige. Chase alignment. Find the place where your weirdest questions are not just tolerated — they’re celebrated. Find the lab where your voice shakes in the beginning —

and becomes thunder by the end. Find the mentor who doesn’t give you answers — just better questions.

Purdue, for me, was that place.

It wasn’t just that the Mechanical Engineering program was top-tier — it was that I could walk into a room and say, “Hey, what if we treated machine learning as a source of epistemic uncertainty in design?” and no one blinked. Instead, they handed me whiteboard markers and said, “Show us.”

Here, you’ll find professors who are giants in their fields and yet leave room for your ideas; students who are brilliant but also kind, collaborative and weird in the best way; and resources that stretch your imagination — and deadlines that stretch your caffeine tolerance.

But most of all? You’ll find space — the kind of intellectual freedom that dares you to chase problems that haven’t been solved, because they haven’t yet been seen from your angle.

So, what would I tell a student?

Choose the place where your curiosity feels invincible. For me, that place was Purdue.

What about the future? What are your goals; what are you looking to accomplish in this field?

If you had asked me this question a few years ago, I probably would’ve rattled off titles: professor, principal scientist, research lead. But now? Now I think in impact, not positions.

I want to build things that live beyond me. I want to shape a future where machine learning isn’t just fast — it’s trustworthy. Where engineers don’t just optimize — they understand the risks, the blind spots, the uncertainties that hide beneath predictions.

In the short term, I want to continue doing research at the intersection of engineering, AI and data science — possibly as a postdoc in computer science, where I can deepen my expertise in model robustness, explainability and real-world deployment of intelligent systems. I’m especially drawn to smart infrastructures, health tech, neuro-inspired control — anywhere that machines touch lives in complex, fragile environments.

In the long term, I want to lead — not by authority, but by originality. I want to create hybrid research environments that fuse engineering, AI, neuroscience, ethics and design thinking — where people like me, who don’t quite fit into one box, can thrive.

But beyond the technical goals, here’s the real dream:

I want to be the kind of scientist whose work makes people feel safer, smarter and more human. I want to teach not just how to build things but how to build wisely.

And in my life? I want to stay curious. Stay uncomfortable. Stay surrounded by people who challenge me to think deeper and laugh louder. I want to write. Mentor. Travel. Maybe one day start a research lab with a ping pong table and a poetry wall. Because why not?

This journey isn’t just about reaching the top. It’s about building a staircase others can climb too.

Might you share with us a little window into your personality: some distinctive trait, habit of mind, hobby/pursuit outside work…?

Let me start with this: I’m the kind of person who writes code all morning and then reads Rumi under a dim lamp at night.

I exist somewhere between logic and lyricism — deeply technical by training, but hopelessly poetic by instinct. I’m an introvert with a wildfire mind — quiet in rooms, loud on paper. I can spend five hours tuning a surrogate model and then another hour staring at the sky, wondering how uncertainty works in people, not just systems.

One of my most distinctive traits? Pattern obsession. I see them in equations, in architecture, in how people speak. I sometimes journal in diagrams. My thoughts arrive as shapes, not sentences. It’s weird. I love it.

When I need to breathe, I vanish into meditation and long runs — usually through places that feel abandoned by time. Old streets. Rainy mornings. Empty parks. That’s where the best ideas sneak in.

I also have a soft spot for animals, cinema and geopolitics — in that exact chaotic order. I’ve had passionate, borderline existential arguments about Christopher Nolan’s timelines. I revere Tarantino — his storytelling, his madness, the rhythm of violence and silence. But above all Hollywood raised me. I may have been born in another part of the world, but my soul grew up in the United States — through film, through stories, through that cinematic sense of possibility.

That’s why I feel I belong here—not just as a researcher, but as a dreamer. I’ve loved the United States of America since my childhood. I can’t deny it!

And somewhere along the way, I picked up a strange habit: turning even small moments into mini research projects. Once, I spent two weeks studying how different cultures design door handles. Don’t ask why. It started as a joke and turned into a deep dive on intuition, touch and the philosophy of interaction. Classic me.

So yeah — I’m a thinker. A builder. A dreamer. I believe in depth over noise, stillness over speed and curiosity over certainty.

And if you’re ever up for a late-night debate about AI ethics, control theory, or whether penguins are secretly introverts too, I’m your guy.