Yes. For a non-CS major applying for product manager jobs, a small product you built and launched yourself is one of the clearest proofs that you can do the work, because it shows you found a problem, made trade-offs and got real people to use something. It helps only if it's real: live, used by people who aren't your friends, and something you can explain decision by decision. It won't get you past roles that require a CS degree.
Does building your own product help a non-CS major get a PM job?
Yes, and it may help a non-CS major more than anyone else. A computer science student has a degree that signals technical comfort. You don't. A live product fills that gap with something more direct: proof that you can take a problem from "people complain about this" to "here's a thing they use."
The people who write about PM hiring mostly agree on this point. David Cai, who went through an APM program and later reviewed hundreds of APM applicants, writes in his account of his own APM resume that the strongest credential for PM ability is having built an actual product yourself. Mika Reyes, a former LinkedIn PM who did not study STEM, argues in her guide to getting a PM role out of college that the best way to show product skills is to start something yourself, and that it counts whether or not the project succeeds. Jackie Bavaro, a PM hiring manager and author, lists an entrepreneurial side project among the "signs of excellence" in her overview of how to become a PM.
Where the advice disagrees, or goes quiet
The sources split on three things worth knowing before you spend a semester building.
Some doors stay closed. Bavaro notes that some APM roles require a CS major or minor. In a Berkeley Haas podcast episode, Yifei Ding, a Haas undergrad alum who became a PM at Instagram, says you don't need a technical background in general, but that some companies may not consider candidates without one. A product won't change a hard degree requirement. Read the minimum qualifications on every posting and don't waste applications on roles that screen you out.
A product is one route, not the only one. Ding got into product through consulting, not a side project. Her point is that product sense, execution and transferable skills from other jobs carry over. Reyes lists internships, freelance work, product reviews and starting an organization alongside building. If you already have strong work experience, a product is a supplement. If you have none, it may be your best evidence.
Hackathons and fake startups count less. Cai says hackathons show interest but carry little weight without follow-through. Bavaro warns against founding a company just to get acquired into a PM job. What counts is real use, not the label.
All four sources are mostly silent on two things that matter now: how to show that your usage numbers are real, and how to talk about a product built largely with AI coding tools. Those gaps are where most students get caught in interviews.
What makes a self-built product count in a PM interview?
A PM interviewer doesn't grade your product like an app store reviewer. They use it as a prompt to see how you think. A product helps you when it gives them material for that:
- A real problem with a named group. "Students in my dorm who split grocery bills," not "people who want to save money."
- Evidence you talked to users first. Who you spoke to, what they said, what you decided not to build.
- Real users who aren't your friends. Even ten strangers using it tells an interviewer more than a polished demo nobody touched.
- Numbers you can back up. Analytics screenshots, a sign-up list, a log. Never round up.
- Decisions and trade-offs. What you cut, what you shipped first, and why.
- An honest failure. The part that didn't work and what you learned.
AI coding tools change one thing here. Building a working app is now much easier than it used to be, so an app on its own proves less. Users, evidence and decisions are what set one candidate's product apart from another's.
What will interviewers ask about it?
Prepare for these. Most come from how PM interviews work in general, applied to your project:
| Question | What they're really checking | Have this ready |
|---|---|---|
| Why this problem? | Can you spot real needs? | Notes from your first user conversations |
| Who uses it, and how many? | Honesty about evidence | Real numbers and where they come from |
| What did you choose not to build? | Prioritization | One feature you cut and why |
| How do you know it works? | Measurement | One metric you tracked and what it showed |
| What would you do next? | Product sense | A specific next step tied to what users did |
| How technical are you? | Can you work with engineers? | Plain explanation of how the product is built, and how you used and checked AI |
The last question is where non-CS candidates gain or lose the most. You don't need to sound like an engineer. You need to explain clearly what each part does, where the data lives, what broke and how you fixed it.
A plan for one semester
- Weeks 1–2: Pick a problem from a group you're part of: a club, a class, a job. Talk to at least ten people who have it.
- Weeks 3–4: Build the smallest version that solves one part of it, and put it online at a public link.
- Weeks 5–8: Get it in front of people outside your friend group. Watch where they get stuck. Fix the biggest problem.
- Weeks 9–12: Track one number, such as weekly users or how many come back, and try one way of reaching new users.
- Throughout: Keep a short weekly log of what you shipped, what users said and what you decided. That log becomes your interview prep and your resume bullet.
Then put the link at the top of your resume, and use it in conversations. Reyes points out that cold online applications rarely work and that alumni referrals and informational interviews do better. A live link gives a PM alum something concrete to look at.
How STARC approaches this
STARC is a small nonprofit, and its one-person company cohort follows much the same plan. Over eight weeks, students from any major use AI to build a small product, talk to potential users, launch it at a public link and look for real users. Each week they write a build log: what shipped, the numbers, what users said, where they got stuck and how they used AI. To count as "Shipped" at graduation, a product needs at least 10 real users who aren't classmates, friends or family, with evidence. There's no certificate. The product, data and logs are the proof.
The first cohort hasn't run yet, so there are no past results to point to. What STARC does and won't do is on the How it works page, and common questions are in the FAQ.
The short version
If you're a non-CS major aiming at product management, build something small and real, get strangers to use it, write down your decisions and be honest about the numbers. It won't beat a hard CS requirement, and it won't guarantee an offer. But for most PM roles, a live product you can explain is the most direct answer to the question every interviewer has about you: can this person actually do the job?
Related questions
Do I need a CS degree to become a product manager?
For most PM roles, no. Several practitioners who hire PMs say non-CS candidates get in regularly. But some companies and some APM programs do ask for a CS or technical degree, so read each posting's minimum qualifications before you apply.
Is a hackathon project enough?
Usually less than you'd hope. A former APM who reviewed applicants says hackathons show interest but count for little unless you kept going, for example by continuing the project or getting real users.
Does it matter that I built the product with AI tools?
Not in a bad way, as long as you say so plainly. A PM interview cares about the problem, the decisions and the evidence, not who typed the code. Be ready to explain how you checked what the AI produced.
What if my product failed?
It can still help. Interviewers learn a lot from a clear account of why something didn't work and what you'd change. A failed product you understand beats a successful one you can't explain.
Want to build your own product?
The OPC cohort: 12 students, 8 weeks, one real product with real users.
Last updated: 2026-10-11. Written by the STARC team. Rules quoted here come from our transparency page; if they change, this post is updated.
More questions
Can I talk about a project I built with AI in a grad school interview?
How to present an AI-built project in a grad school interview: what faculty will probe, how to describe what AI did, and a 2-minute answer structure.
Do students from less well-known universities stand a chance in a blind-reviewed program?
What a blind review removes, what it can't, and how a student from a less well-known university can build an application that wins on the material itself.
What's the difference between a student one-person company program and an AI side hustle course?
A one-person company program has you ship a product to real users; an AI side hustle course sells you lessons. How to tell them apart before you pay.