Building A Simple Model on a Quantum Computer
Introduction
I got into quantum computing a few years ago when I joined a fellowship program called QuForce that connects experts in quantum computing with professionals or student learners. I was paired with a mentor and I wrote a white paper about how companies like IBM might offer quantum computing as a cloud-based service to customers in the future. The paper drew from my background as a product manager in cloud infrastructure and security, and I attempted to draw parallels between how cloud computing was distributed as a service and how quantum and classical computing might be distributed together as a service. Here’s the paper for anyone interested:
A few years later, during NYC Tech Week in 2025, I attended an event hosted by IBM at their offices about quantum computing. Here, I got my first exposure to Qiskit — IBM’s open-source quantum computing SDK — and how it can be used to model and solve quantum problems over an infrastructure that leverages both classical and quantum computers.
I was fascinated, to say the least. I mean, running a simple model on a real quantum computer felt like I was a part of something huge. Something that wasn’t yet perfect, but was soon going to be monumental to our future. In talking to some of the other attendees here and at the NYC Quantum Computing Meetup community, I learned that quantum computing didn’t need to be error-free in order for it to be useful. More specifically, quantum modeling—leveraging quantum principles on classical computers (and sometimes quantum computers) to simulate solutions to real problems— was something that already existed. In fact, some of the larger quantum computing companies are selling simulation systems and expertise to do exactly this. Platforms like PennyLane and companies like Google are building out Quantum ML or QML, a field of study built around quantum modeling. And even more interestingly, these companies have paying customers. Research teams across finance, health, government, and more are willing to pay for quantum modeling expertise and systems.
This is why I’m starting this writing series. I believe we’re at a turning point in quantum computing. 2026 could be a pivotal year for more enterprises and their leaders to start thinking about quantum modeling. The hardest part of the quantum modeling process is translating a real-world problem into a quantum problem, and that’s the opportunity, challenge, and skillset I’m excited to explore and shape here.
Get Alaap Murali’s stories in your inbox
Join Medium for free to get updates from this writer.
Starting with the Magic 8-ball problem.
The Magic 8-ball problem
This problem is a great introduction to IBM’s Qiskit SDK, and to quantum computing in general. The problem is to generate a random response from a digital Magic 8-ball, one of those spherical toys you probably saw as a kid that “predicts” the future when you ask it a silly question. More technically, it’s effectively a random number generator — but unlike classical computers that simulate randomness, I used Qiskit to run my code on a real IBM quantum computer, making the result truly random. Pretty neat! Below is the project on my GitHub. Each code block is commented so you can follow along and understand the core components. Thank you to the IBM Quantum Education team for the exercise and structure.
It’s worth mentioning that I’m supplementing these coding projects with Thomas Wong’s book Introduction to Classical and Quantum Computing. Highly recommend!
Applying this method to a broader class of problems
While the Magic 8-ball problem is a pretty simple problem, it sets the foundation for understanding where quantum computing is actually useful and how SDKs like Qiskit simplify it. As I study quantum modeling further, my hope is to be able to construct new problems using some of the same Qiskit components and methods I used here.
While circuits, algorithms, and the underlying math might get more complex, my goal is to build an intuition for modeling problems as quantum problems. And crucially, I’ll be thinking about breaking up real world business problems into classical and quantum components: after all, we will live in a future where the two work together.
