SMART CITIES · AI · SUSTAINABILITY

Willy Tai

Ten years delivering technology in regulated environments. Now working on smart estates, AI and sustainability, with demos you can run.

I deliver smart-estate technology in regulated environments, and show it working.

A S$35M command centre with a digital twin for a Government agency. The first of Singapore's new digital banks to go live. More than fifty facilities instrumented. The record is on this site, and so are 14 working demos built to the same rules.

Try a digital twin you can operate

Screenshot of the Estate Twin: a 3D block with lit sensor points and a work-order panel
FEATURED · LIVE DEMO · SYNTHETIC DATAEstate Twin. A 3D twin of a small estate. Inject a fault, watch it reach the floors that machine serves, and approve the work order.Open full page →
SHOW
Screenshot of the Estate Twin: a 3D block with lit sensor points and a work-order panel
Smart Cities· Digital twin

Estate Twin

A 3D twin of a small estate. Inject a fault, watch it reach the floors that machine serves, and approve the work order.

A fault photo with an annotation box and the proposed work order beside it
AI· Smart FM

Snap to Work Order

A fault photo becomes a work order. The model proposes the fault class, rules set priority and SLA, and a named person approves.

Air-handler residual charts with a diagnosed component
Sustainability· Real data

Plant Diagnostician

Air-handler and rooftop-unit faults diagnosed in public test data from two US national labs, without the answer key, then scored against it.

Process map of maintenance work orders with a parts-wait bottleneck highlighted
Smart Cities· Process mining

Work Order Miner

850 synthetic work orders mined into the as-is maintenance process. The CMMS reports 93.8% SLA compliance; the clock says 73.4%.

Schematic of a chiller plant with pumps, cooling tower and chilled-water loop
Smart Cities· Explainer

Chiller Plant Anatomy

Six ways a building is cooled, each drawn as a working schematic. Hover any pump or valve, run the start-up sequence, drag the building load.

Device registry table with lifecycle findings
Smart Cities· IoT lifecycle

IoT Fleet Registry

124 synthetic devices checked against seven lifecycle rules, with a risk register, a twelve-month calendar and an onboarding gate.

Alarm list reduced to the ones needing a person
Smart Cities· Alarm management

Alarm Referee

1,157 synthetic building alarms in one week reduced to the 29 that need a person. Two cascade storms traced to their source.

Data profiling screen with quality gates
AI· Data onboarding

Day-One Data Room

Three source systems that disagree, profiled, quality-gated and reconciled, with a person deciding the two uncertain merges.

Acceptance criteria checklist with hashes
AI· Acceptance

Go-Live Gate

Five acceptance criteria fixed and hashed before any result is seen. Week-six drift breaks the signed operating point and withdraws the approval.

Vendor scorecard with evidence gaps
AI· AI procurement

AI Vendor Scorecard

24 due-diligence questions where a claim without evidence scores at most a third. The sample vendor claims 77% and evidences 31%.

Climate disclosure checklist with assurance scores
Sustainability· ESG assurance

Climate Disclosure Bench

A twelve-requirement climate checklist where every verdict must quote the report or abstain, re-scored the way FY2029 limited assurance will.

A chart produced from a plain-English question
AI· Data access

Talk to Your Data

Type a question and get a chart from a defined metric catalogue, with provenance shown and ambiguous terms sent back for clarification.

Generated SQL beside its semantic check
AI· Text to SQL

SQL Trust Gate

Generated SQL checked against a semantic contract before it runs. Where the model abstains is shown next to where it would have returned a confident wrong number.

02Article series

All articles →
Series · four parts · complete

The Minimum Twin

Four articles on how little digital twin an estate needs to run its monitoring loop. It starts with a twin you can operate in the browser, removes most of it, rebuilds it from one photograph, and ends with what an owner should put in a tender.

Start with Part 1
  1. 01A twin you can operate
    A live twin in the browser.
  2. 02State, kind, rough location
    The Lattice view removes most of the model and still catches the fault.
  3. 03A twin from one photograph
    The twin's geometry taken from one photograph through a table that marks every value observed, inferred or assumed.
  4. 04What to put in a digital twin tender
    Which part of a twin budget buys the twin, and which buys polygons?
SURBANA TECHNOLOGIES · 2022 TO NOW
Joint programme manager for a S$35M M&E technology refresh for a Government agency: an integrated command centre with a digital twin, more than ten suppliers, data from BMS, power SCADA, security and fire systems.
SURBANA TECHNOLOGIES · 2022 TO NOW
Led a S$4M smart-facilities refresh across 50+ facilities: BMS renewal, IoT sensing, a microgrid and an FM command centre. Separately, a S$4M green-technology programme featured in The Straits Times and The Business Times.
GREEN LINK DIGITAL BANK · HONG LEONG FINANCE · PROTIVITI
Pioneer-team project manager at the first of Singapore's new digital banks to go live, carrying a S$15M+ portfolio under MAS TRM. Before that, a three-year IT transformation at a listed finance company, and audit and risk advisory.
01 · CLIENT DELIVERY

Client-facing delivery

I scope the problem in the client's language, deliver the build, and land it as a business number. Ten years of the record above happened in front of clients: agencies, banks, a retail conglomerate, dignitaries touring a command centre.

02 · CAPABILITY BUILDING

Building the internal practice

Adoption playbooks, governance processes, and the evidence infrastructure that defends both. This site is the reference implementation, built end to end by one person, so I know where every hour and every risk sits.

03 · TEACHING

Teaching while building

Teaching practising accountants through ISCA since 2019, supervising capstone projects at Nanyang Polytechnic, an ACTA adult-education certification, and a PhD on IoT management in smart buildings completing in 2027.

“AI should make a human's judgment better, or replace only the decisions that were never judgment to begin with.”

Get in touch