Chapter 05
The Convergence
2025–Present
The developer and the PM finally merged — managing pharma capital projects while building AI agent systems that reinvent how project managers work.
Auckland
New Zealand was a dream. The nature, the beaches, the mountains, the people, the culture. I moved here first, then found the role at Douglas Pharmaceuticals.
After thirteen years building digital products across five countries — Singapore, Myanmar, back to Singapore, and now New Zealand — something clicked. Every chapter had been preparation for this one.
Douglas Pharmaceuticals
Douglas Pharmaceuticals is a pharmaceutical manufacturer operating GxP-regulated manufacturing and quality control laboratories. My role is Digital Project Manager, managing capital projects in digital and IT systems.
The projects span laboratory instrument systems, building management and environmental controls, packaging compliance and serialisation, and cross-functional coordination across QA, manufacturing, IT, and operations teams.
Pharmaceutical manufacturing is one of the most regulated environments you can work in. Every system change requires validation. Every validation requires evidence. Every piece of evidence requires traceability. It's the most demanding application of project management discipline I've encountered — and it uses everything I've learned.
This is where the earlier chapters pay off:
- The Code gave me the instinct to understand technical architecture — I can read a system diagram and see where the risks hide
- The Bridge gave me stakeholder fluency — I can translate between QA's compliance concerns and IT's implementation constraints
- The Scale gave me team-building experience — I know how to coordinate cross-functional groups toward a shared outcome
- The Systems gave me enterprise complexity tolerance — I can hold multiple concurrent projects with interdependencies without losing the thread
The AI ecosystem
Outside of Douglas, I've been building something else entirely.
Over the past year, I've constructed an AI agent ecosystem — a personal project that reimagines how project managers work. It's the developer in me and the PM in me, finally working on the same problem.
The idea is simple: a project manager's day is filled with meetings, decisions, documents, communications, and context-switching. Most of that work involves gathering information, synthesising it, and acting on it. What if an ecosystem of specialised agents could handle the gathering and synthesising — so the PM can focus on the acting?
That's what I built. Not a single chatbot, but an interconnected system where each piece does one thing well and passes context to the next:
- An agent that processes meeting transcripts into structured summaries, extracts action items, and updates the project knowledge base — automatically
- An assistant that prepares meeting briefs by pulling relevant history, open risks, and pending decisions from across all my projects
- A communication drafter that adapts tone based on the recipient — formal for vendor escalations, concise for engineering updates, structured for stakeholder reports
- A knowledge engine that remembers everything — every decision, every correction, every observation — and makes it searchable months later
The pieces connect through a shared memory layer. When the meeting agent processes a transcript, the insights it extracts become available to every other agent. When I correct a communication draft, that preference gets remembered for next time. The system learns from its own outputs.
The most interesting thing about building this system isn't the AI. It's that every design decision draws directly from PM experience. The agents are narrow because I've seen general-purpose tools fail. They show their work because I've learned that trust requires transparency. They embed into existing workflows because I know people won't change how they work just because you gave them a tool.
What makes this a convergence — not just a side project — is that it couldn't exist without every previous chapter. The developer from Chapter 01 can build the system. The PM from Chapters 02–04 knows what the system should do. Neither skill alone would be enough.
Why "convergence"
I didn't plan this arc. Each chapter felt like a departure at the time — leaving code for management, leaving Myanmar for Singapore, leaving startups for enterprise. But looking back, every move added a capability that the next chapter required.
The Code taught me to build. The Bridge taught me to translate. The Scale taught me to grow. The Systems taught me to hold complexity. And now, in The Convergence, all of those skills fire simultaneously — managing regulated capital projects during the day, building an AI ecosystem that makes that work better at night.
This isn't a career pivot. It's a career convergence. Every chapter was a thread. This is where they weave together.