Projects
Things I've built, studies I've run, and ventures I've tried to get off the ground.
Production systems
AI Engineer Intern, Garuda Robotics
You can delete a bad answer. You can't un-fly a drone. That changes where the checks go.
- Agentic AI
- LLM Safety
- Systems Design
- Drones
Data Science Intern, SP Digital
My first system where someone was paid to break it. That changed how I build.
- LLM Safety
- Guardrails
- Evaluation
- Langfuse
- LangGraph
AI Architecture Strategy Engine
Personal project
A tool for deciding whether to use prompting, retrieval, or fine-tuning for a given job. Most teams pick by instinct or by whatever they read about last. But the costs are all knowable, so I built something that works it out: it scores each option against your budget and latency limits, runs a thousand simulations with the numbers nudged around, and tells you where the answer flips from one to another.
- Python
- Multi-Agent
- LLMs
- System Design
Socratic Digital Twin
Developer, NUS Faculty of Dentistry × School of Computing
A collaboration between the NUS Faculty of Dentistry and the School of Computing. It's an AI tutor for dental students that's built to refuse to answer. The subject is orthodontic clinical reasoning, where being handed the answer defeats the point, so the system asks questions back instead. That constraint drives everything: a multi-stage pipeline that decides what to ask next, retrieval over the faculty's own teaching material rather than the open web, and a review step where a clinician signs off on content before a student ever sees it. Currently in development.
- LangGraph
- RAG
- Postgres
- Clinical AI
PII Redaction Evaluation
Data Science Intern, SP Digital
At SP Digital, a separate system strips personal information out of customer call transcripts. I built the evaluation pipeline that measures how well it does that. Most of the work wasn't the measurement, it was deciding what counts: a receipt number isn't personal information, a partial email address probably isn't either, and those rules have to be written down and applied consistently before any score means anything. I scored it so that missing something counts as worse than being over-cautious, because those two errors aren't equally bad here.
- PII
- Evaluation
- NLP
LLM Evaluation Framework
Personal project
A tool for comparing language models on the same task and seeing what each one actually costs you in quality, money and speed. I ran it across three different kinds of work: summarising lectures, reasoning through business decisions, and ranking documents by relevance. The most useful thing it turned up was a measurement problem. The standard ways of scoring text similarity rate one model far worse than another purely because it wraps its answer in formatting, when both are ranking the documents equally well. For anything where the output has a structure, those metrics quietly mislead you, and you need one that measures the thing you actually care about.
- Evaluation
- Python
- Benchmarking
MarkBind — Open Source Contributions
Contributor
Contributions to MarkBind, an open-source documentation site generator maintained at NUS. I was picked for it off the back of the software engineering course. It was the first time I'd worked in a codebase I hadn't written any of, with a review process I had to satisfy.
- Open Source
- Java
- Documentation Tooling
TrackUp
Team project
A desktop contact and event manager for founders and small business owners, built as a team software engineering project. The deliberate choice in it is that everything is driven by typed commands with a graphical view alongside, rather than the other way round. That's the opposite of what most contact tools do, and it's right for the specific person who lives in a terminal and finds clicking through forms slower than typing what they want.
- Java
- JavaFX
- CLI
Research
Independent study
I wrote down the method before I looked at any data. Halfway through I found what I was testing for, then worked out the maths was producing it, not the market. Three months later, the result held on data nobody had seen.
- Market Microstructure
- Pre-registration
- Statistics
- Python
Independent study
The model worked, but the metric was worthless and the monitoring was blind. A study in how machine learning projects fail even when the model itself is good.
- Machine Learning
- Evaluation
- Pre-registration
- Causal Inference
Singapore Society Simulation
Researcher, NUS Odyssey
Can AI agents stand in for real people when you want to know how a population thinks about something? I gave each agent a demographic profile and had them argue a policy question, then compared what they concluded against what real people had said online. They came out quite different. Changing who was in the simulation also moved the answer a long way, which means you'd be measuring your own choice of participants as much as anything else.
- LLM Agents
- Social Simulation
- Research
Ventures
Lecture AI
Case study
Counter: pricing intelligence for Singapore cafes
Co-founder
Counter shows a cafe owner how their menu prices compare with the cafes around them. You upload your menu as a PDF or a photo, the AI reads it, you pick your competitors on a map, and you see where each item sits against theirs, with the source and date of every price. You can try a new price on an item before changing anything, and see new items and offers at nearby cafes. It runs on a live benchmark of more than 6,000 menu prices from over 100 Singapore cafes. The Weekly Bake, our newsletter, is how we reach owners. I'm building the Counter demo and adapting the landing page, working with three teammates from my NUS venture course. We started with food trend tracking for operators and moved to pricing in October. It's still early: we've interviewed owners in person and put a paid waitlist on the landing page to see who commits. One lesson from the first version stuck: you can't sell a monthly subscription into an industry where most shops never turn a profit.
- Product
- AI
- Market Intelligence
- F&B
Knocks
Solo build
A card game my friends and I have played for years, which existed only as a physical deck, so it needed everyone in the same room. I built the online version. It then spread by word of mouth to friends of friends I've never met, who kept playing it. Nothing about it was technically hard, and it's still the first thing I built that reached people I didn't tell about it.
- Real-time
- Multiplayer
- WebSockets
Sixer
Solo build
A cricket draft game we used to run over WhatsApp, badly. Group chats are a terrible place to hold a game with rules. Someone always has to arbitrate, someone always misses a turn, and the state of play lives in whoever scrolled back furthest. So I moved it somewhere that actually holds the rules and keeps everyone in sync.
- Real-time
- Multiplayer
- WebSockets
Earlier work
ChessPhere
Co-founder
I've played competitive chess since I was young, and when the pandemic shut down every over-the-board tournament, the thing that disappeared wasn't the game. Online chess was fine. What disappeared was the community around it. So a few friends and I started running virtual tournaments and workshops, and we were drawing over a hundred players to each one. It never made money and it was never going to. What it taught me is that a small niche can matter enormously to the people inside it, and that's a reasonable thing to build for.
- Community
- Events
- Chess
Donation Nation
Co-founder
This started with me giving away things from my own house to local NGOs during the pandemic. It worked, and it obviously didn't scale, because the bottleneck was one person with a car and a limited amount of stuff. So we built a platform connecting donors with communities that needed things, working through established NGOs and logistics partners rather than moving goods ourselves. The hard part was never the donating. It was that nobody could find each other.
- Social Impact
- Platform
- NGO Partnerships