Too busy to watch the full episode? Here's the short version — James Liu's journey from SWE intern to full-time PM at Crypto.com, and what he learned about domain expertise, knowing yourself, and why onboarding funnels are the most important thing nobody talks about.

Product management has long lived in the shadow of software engineering in CS circles — seen as the "softer" path, or worse, not a real path at all. But for James Liu, a fresh NUS Computer Science graduate now working as a full-time Product Manager at Crypto.com, the road from code to product wasn't a detour. It was the whole point.

In this episode of Lorong Product, we sat down with James — an NOC Silicon Valley alumnus, former ByteDance PM intern, and self-described strategy game enthusiast — to trace his career from year-one SWE internships all the way to managing new user activation at one of the world's largest crypto platforms. Along the way, he shared hard-won lessons on landing PM internships, navigating PM crises, and why knowing yourself might be the most underrated skill of all.

1. You Don't Have to Commit to SWE (or PM) From Day One

James's journey didn't start with a grand plan. It started with taking the first initiative he could get — a SWE internship at a fintech startup called Pentas in year one, on the recommendation of a mentor.

But something unexpected happened. Beyond the frontend work and automation testing, he noticed the strategy behind the product — how decisions were made, how the company tried to grow, how a product reached the average consumer. That curiosity planted a seed.

"I realised it's quite interesting, not just the software engineering part, but actually how to grow a company, how to get it from their side to the average consumer."

By year two, he was doing frontend work at EY. By year three, he had applied for NUS Overseas Colleges (NOC) Silicon Valley — the NUS programme that lets students intern at overseas startups. He applied to an e-commerce company. They told him the software roles were gone. So he joined as a PM instead.

What's notable here is that James didn't abandon engineering — he used it as a foundation. His technical fluency later became an asset in ByteDance interviews, where system design and technical knowledge still came up. He also recommends CS2109S (Intro to AI and Machine Learning) for the AI literacy it builds, and CS3216 (Software Product Engineering for Digital Markets) for its rare combination of building, marketing, and actually growing a real product.

3. Startup vs. MNC: Offence vs. Defence

One of the sharpest frameworks James shared was how PM work differs fundamentally between startups and large companies — not in title or tools, but in mindset.

At Trendsi (startup), the goal was simple: attract as many users as possible. Churn didn't matter as much — growth did. One initiative he ran was a "happy event trigger" for app reviews: instead of prompting users randomly to rate the app, the team tested asking immediately after a successful purchase — the moment users were most satisfied. The result? App reviews tripled.

"We realised that the happiest time is when they've just finished a purchase, which makes sense. So we just did that."

At Crypto.com (MNC), the calculus flips. The marketing team might be paying to acquire 500,000 users a month — so the PM's job isn't to find users. It's to make sure those users don't drop off before they activate. Brand reputation is hard-won and worth protecting.

"At an MNC you're playing defence. At a startup you're playing offence."

This distinction also applies to the type of PM work James did at ByteDance, where he was building AI pipelines and writing PRDs for LLM and RAG systems — more systems-thinking, more analytical. Versus growth-oriented PM work at Trendsi and Crypto.com, which is more about user psychology, funnel optimisation, and behavioural nudges.

His take: if you want to build from 0 to 95, go to a startup. If you love the analytical, optimisation-heavy work of going from 98 to 100, big tech might suit you better — and a CS background is genuinely useful there.

4. Know Your Domain, Know Your Users, Know Yourself

When asked about essential traits for PMs, James resisted the easy answer (extroversion, communication, leadership). Instead, he offered three things he believes matter most:

Know your domain. Even at the senior level, companies don't hire by years of PM experience — they hire by domain. Fintech payments. B2B platforms. Consumer growth. The more specific your expertise, the more valuable you are.

Know your users. Genuine empathy with users isn't a soft skill — it's a competitive advantage.

"If you're not a frequent user of the product, it's very hard to come up with good ideas. If you don't use Reddit, it's hard to know how to improve Reddit."

Know yourself. This one surprised him the most, but it's become a core part of how he advises students. Your competitive edge as a PM comes from your life path so far. Were you a sports captain? Aim for programme management or leadership-heavy roles. Have a technical background? AI and data PM roles. Ran a small business? Growth PM is a natural fit.

"A lot of people overlook this, but knowing yourself first is quite important to choose what will suit you the best."

5. How to Navigate PM Crises Like a Pro

In a new segment, James was tested with two real Reddit PM dilemmas. His responses revealed the kind of structured, stakeholder-aware thinking that separates good PMs from great ones.

Scenario 1: A technical PM is drowning — spending 15% of their week on bug fixes, managing two modules, and being asked to work weekends by a manager who wants faster output.

James's advice: quantify everything before escalating. Don't just tell your manager you're stretched. Show them that the 15% bug-fixing time represents a 30% risk of system failure. Show them the estimated revenue from the subscription revamp. Put both on a page, side by side, and ask your manager to weigh in on the tradeoff.

"You give him a recommendation, but make sure he also has some say in it. There's no right or wrong in prioritising — you just make sure everyone stays in the loop."

Scenario 2: A senior PM on a high-status LLM project feels invisible — no agency, only tactical work, unclear ownership among too many PMs.

For junior PMs and students, James's advice was clear: always optimise for learning. Go where you have agency to shape decisions, write solutions, see results, and build product sense. That's irreplaceable.

But he also added nuance: before assuming it's a structural problem, diagnose the root cause. Are your solutions not landing? Are you not communicating the right way? Fix that first. If it really is structural — if the company genuinely doesn't empower its PMs — then make a move, whether lateral or external.

"Read the Glassdoor reviews before you join. See what you can learn. If a lot of them say it's toxic or they don't give you scope until you're more senior, I don't think it's a great learning environment."

6. The Hidden Weapon: Soft Skills (Specifically, Flexibility)

For his final reflection, James was asked what skill matters most for PM but students fail to train. His answer wasn't data. It wasn't SQL. It wasn't even communication in the broad sense.

It was flexibility.

He pointed to IT2900, a course at NUS that simulates corporate interpersonal dynamics and builds emotional intelligence. But the real lesson came from the job itself: different stakeholders require completely different approaches. Some respond to directness. Some need to be coaxed. Some will ignore you until you apply gentle pressure at just the right moment.

"You need to be flexible in your approach — more individualistic. Some team leads, you softly put your request, your ego aside, and that works. Others, you wait a week and realise nothing's moved, and only then do you nudge."

He also made a point that stuck: in NUS, CS students tend to interact mostly with other CS students. Same classes, same interests, same thinking patterns. Breaking out of that bubble — talking to people from other faculties, understanding how they think and what they care about — is one of the most valuable things a future PM can do.

Conclusion

James Liu's journey is a useful reminder that the PM path rarely announces itself. It often starts with a curious SWE intern noticing there's something interesting beyond the code. It grows through small businesses, failed gift boxes, referrals from seniors who noticed good work, and a willingness to try things before knowing if they'll stick.

What ties it all together isn't a perfect portfolio or a prestigious company name. It's the habit of knowing your domain, understanding your users, and being honest about who you are and what you bring to the table.

Whether you're a CS student wondering if PM is for you, or a SWE intern who keeps thinking about the product side — James's story shows that the best career moves aren't always the most obvious ones. But they're almost always worth taking.

Watch the Full Episode

Click below to watch the full conversation with James Liu on Lorong Product.

https://youtu.be/4sskNaENOjI

Enjoyed this? Check out our other episodes for more honest conversations about careers in product, design, and tech.

About the Author