Shopee Cuts Developer Jobs Because of AI?

ATP is presented by:

Welcome back to The Asia Tech Podcast. Today we discussed Shopee cutting developer jobs because of AI, what it really means when a company bets on smaller engineering teams, how AI voice platforms are personalizing customer service at massive scale, what Zoom thinks the future of workplace productivity looks like, vertical AI transformation, the looming threat to your encryption, navigating the noise in AI-powered sales, API-first investing, and what 10,000 people at a single AI conference says about where the industry is heading.

We were joined by:

• 23:25 Roshan Gill - Country Manager, Malaysia at AI Rudder

• 43:06 Carlos Quaderi - Head of Asia at Zoom

• 1:03:04 Touchapon Kraisingkorn - Co-founder, Executive Director and Chairman of AI Research & Application Center at Amity

• 1:22:56 Samuel Tseitkin - CEO at ExeQuantum

• 1:42:24 Serena Lam - Founder & CEO at Fuzzy AI

• 2:07:30 Chris Birrell - Founder and Managing Partner at Func Ventures

• 2:23:30 Peter Noszek - Co-Founder at SuperAI

Here is an overview of the topics we discussed today:

Shopee’s parent company Sea posted $7.1 billion in revenue in Q1 2026, up 46.6% year on year. So why is it cutting 8% of its developer workforce? The story isn’t financial stress — it’s a structural bet on a different kind of team. The idea: one great developer managing a fleet of AI coding agents can now outperform what used to take five people. QA roles and routine coding are the most exposed, especially work that is repetitive and doesn’t require original thinking.

But here’s the part nobody is saying out loud: is this bet actually going to pay off? The CTO of Uber already admitted they burned through their entire AI coding budget faster than expected. Pulling more productivity out of AI agents requires a rare kind of talent — someone who can direct, review, and govern what the AI builds. Are there enough of those people out there? And what happens to companies that cut too fast and find out the answer is no?

Roshan from AI Rudder explained how conversational AI now does far more than field calls. Before a customer picks up the phone, his platform builds a profile based on past behavior — what products they asked about, when they usually call, what they’re likely to need next. By the time the call starts, the AI already has a plan. And when a human agent is in the loop, they get real-time suggestions on screen so they’re never scrambling. What does this mean for what customers should actually expect from service interactions going forward?

Carlos from Zoom made a case that the real problem in most businesses isn’t a lack of tools — it’s too many. AI usage on Zoom tripled year over year, and the reason, he believes, is that they’re focused on a different model. Instead of a prompt box, the conversation itself becomes the trigger for action. Whatever happens in a meeting should automatically move through your other tools and land as a real output, without you switching apps. Could the way we talk eventually replace the way we click?

Samuel from ExeQuantum raised a timeline that should be on more people’s radar. The encryption protecting your bank transactions, your messages, and essentially everything digital today relies on math problems a quantum computer will eventually solve. Gartner and Google both put that window around 2029 — starting with well-funded state actors going after specific high-value targets. The fix exists: post-quantum cryptography, a new class of encryption built to survive quantum attacks. But most organizations haven’t started migrating. How much runway is there, really?

Serena from Fuzzy AI named something a lot of salespeople feel but rarely say: AI has made the inbox worse, not better. Her platform tries to cut through that by helping B2B teams research smarter, personalize more thoughtfully, and build the kind of trust that makes outreach actually land. The goal isn’t to replace who you are — it’s to make the tools simple enough that you just tell them what you want, and they figure out the rest. What does trust-first selling actually look like when everyone has the same AI tools?

Peter from SuperAI described a conference that brought together Chinese open-source model teams, Silicon Valley labs, and Southeast Asian companies in Singapore. What stood out wasn’t the size — it was the diversity. AI isn’t a single-origin story anymore. The gravity isn’t shifting from West to East so much as becoming genuinely multi-polar. And Singapore keeps emerging as the neutral ground where those worlds connect. What does it mean for the region when the biggest conversations in AI are happening here?

There’s more in the full episode — Chris from Func Ventures on why he built what may be the world’s only API-first venture fund, and Touchapon from Amity walking through exactly how they acquire companies and rebuild their products using vertical AI expertise.

Watch the full episode here:

Watch the episode

Twilio is the secure and scalable platform to build amazing customer engagement without the complexity, with the AI-driven workflows you need to connect.

Visit Twilio.com to start building today

Boomi, the data activation company for AI, powers the agentic enterprise by bringing data to life across the business. Trusted by over 30,000 customers and supported by a network of 800+ partners, Boomi helps organizations of all sizes achieve agility, efficiency, and innovation at scale.

Visit www.boomi.com to learn more