I've worked as a product engineer for over seven years, and I mentor developers on ADPList. I still remember how it felt when every new concept or abstraction seemed huge, so in every session I try to explain things in simple words.
After a lot of sessions, I noticed something: the same five questions keep coming back. So here they are, with the answers I give, in case they help someone before their first session.
1. "What does a product engineer do, exactly?"
A product engineer writes code with the user in mind, from the first idea to the moment real people use the feature.
In practice, that means a few habits on top of solid engineering:
- We join the conversations about why a feature matters, alongside design and product.
- We suggest a smaller first version when it reaches users sooner.
- We build across every layer the feature touches, from the UI to the API.
- After the release, we look at the numbers and talk to users to see whether the feature solved the problem.
Here's how that looks with one feature. The team wants a CSV export. Before building anything, we ask who needs it and why, and learn that finance pulls the numbers every Monday morning. So we release a simple export with the three columns finance uses, in two days. A week later, we check the usage, see the downloads every Monday at nine, and suggest a scheduled email as the next step.
That last habit defines the role best. For a product engineer, the work continues after the merge, and the real finish line sits in the user's day.
It also explains why product engineering interviews ask so many "why" questions, and we come back to that in question 4.
2. "Should I go front-end or full-stack?"
Both paths are great, and they ask for different things.
Front-end roles usually expect depth: UI and UX details, accessibility, performance, and often design systems. Full-stack roles expect breadth: comfort with APIs, databases, and deployment, on top of the UI.
A practical way to decide: we open twenty job postings for the role and city we want and count what they ask for. The market tells us more than any opinion online. Then we start from the side closest to our current strengths and grow from there.
Every skill we already have matters, by the way. A former designer brings an eye for UI, a former teacher explains things well, and a former support agent knows exactly where users get confused.
3. "How do I stay confident when everyone around me seems to know more?"
Almost everyone in tech asks this at some point, including people with many years of experience.
What helps most is changing what we compare. We tend to hold our own behind-the-scenes, with every doubt and every bug, next to other people's highlights in standups and on LinkedIn. That comparison feels real, and it measures two different things.
Three habits that help:
- Keep a "done" list. Every Friday, we write down three things we learned, fixed or helped with. After a month, the list shows the growth our memory skips over.
- Ask early, and ask well. A question after fifteen minutes of trying saves hours, and a clear question shows how we think. A simple shape works every time:
"I'm working on the login redirect. I expected the dashboard to open after login, and the home page opens instead. I've checked the route config and the session cookie. Which direction would make sense to look next?"
- Teach what we learn. When we explain something to a person newer than us, in a mentoring call, a blog post or a short demo, we see how much we already know.
We build confidence from evidence, and these habits help us collect it week by week.
4. "How do I prepare for a product engineering interview?"
Product engineering interviews check two things: whether we can build, and whether we care about the people using what we build.
My advice is to prepare three stories before anything else:
- A feature we took from idea to real users
- A bug that took a while to find
- A disagreement with a teammate, and how we solved it
For each story, we write down the situation, what we did, the result, and what we learned. Then we pick one key message per story, the single point we want the interviewer to remember.
Here's a sample story in that shape, ready to adapt:
Situation: Our checkout page took four seconds to load, and users left before paying.
Action: I profiled the page, found a large date library loading on every visit, and replaced it with the browser's built-in date formatting.
Result: The page loaded in a little over one second, and completed checkouts went up the following month.
Lesson: I measure before I optimise, because my first guess pointed at the images.
Key message: I fix problems with evidence.
And we practise out loud. Answers that sound great in our heads often come out three times longer when we say them. A mock interview with a friend, or with a mentor, shows that in ten minutes.
5. "How do I handle a conflict with a teammate?"
Early, and in private. Small tensions grow fast in group channels, and a calm one-to-one solves most of them.
The trick is to act as the facilitator of the conversation, even when we're one of the two sides. A few phrases help at each step.
To open the conversation:
- "Can we take ten minutes to talk about how the release went? I'd like us both to feel good about the next one."
- "I want to understand how this looks from the other side before I share my view."
To listen:
- "Let me check that I got it right: the main worry is the deadline, and the scope question came later?"
- "What would a good outcome look like here?"
- "What else should I know?"
To move from positions to needs:
- "What matters most here?"
- "It sounds like we both want a stable release. Where exactly do we differ?"
To disagree kindly:
- "I see it differently, and here's why."
- "Can we try both ideas on a small piece and compare?"
To close:
- "So the next step is a shorter review checklist, and we check in on Friday. Does that work?"
- "Thank you for talking this through."
These phrases use "I" and "we" throughout, and that keeps both people on the same side of the table, facing the problem together.
One bonus for code reviews: we label comments clearly. "Nit" marks a preference, and "blocking" marks something that needs a change. That one habit removes a lot of friction before a conflict even starts.
What would we ask a mentor?
These five questions cover most of my sessions, and every mentee adds a new angle.
So what's the one question we'd love to ask a mentor right now? Drop it in the comments, and I'll answer as many as I can, right here in the thread. 👇
If a longer conversation would help, I'm on ADPList.
Top comments (0)