Aditya Tri Kurniawan

Director & CTO — PT Machvory Digital Engineering

I build factory systems end to end: the sensor board and its firmware, the Modbus or OPC UA link that gets the signal out of the machine, the time-series backend behind it, and the screen a plant manager opens in the morning.

I make factory data legible

A machine on a production floor is usually silent about itself. It runs, it stops, it draws power, and nobody can tell you how much of any of that happened last Tuesday. My work is closing that gap — from putting a sensor where there was none, to showing the result to someone who can act on it before the cost is already spent.

Selected work

What I have built

Every screenshot below is captured from the running product. No mockups, and no numbers we cannot show the arithmetic for.

Platform · 7 capabilities

TNS — Operational Performance & Energy Intelligence

TNS answers whether the machines are actually working, and what they cost to run. OEE per machine and per shift without anyone filling in a spreadsheet. Stoppages detected and sorted by cause. Energy metered down to the individual machine, including kWh per unit produced — the number that quietly tells you a line is degrading.

Energy Monitoring Energy Efficiency OEE Downtime Production Digital Twin Mobile

TNS energy monitoring dashboard — Energy monitoring — consumption, cost, CO₂ and peak load against the contracted limit.

TNS OEE dashboard with availability, performance and quality gauges — OEE per machine and per shift, computed from machine signals.

TNS 3D digital twin of the production floor — The production floor as a 3D model, running on the plant's own CAD geometry.

TNS downtime monitoring with stoppage causes — Downtime — when a machine stopped, for how long, and which cause it belongs to.

My role: Product architecture, connectivity layer, data model, dashboard design
Stack: TypeScript · React · Node.js · time-series storage
Field inputs: Modbus RTU/TCP · OPC UA · MQTT · power meters · PLC

Platform · 6 capabilities

OPTIMUS — Predictive Maintenance & CMMS

Where TNS asks how well a machine is running, OPTIMUS asks how long it will keep running. It watches vibration, temperature and current on assets a plant cannot afford to lose, flags the change early, and turns that finding into a work order with a schedule, a checksheet and a spare part attached — not a note in someone's phone.

Condition Monitoring Predictive Maintenance Asset Health CMMS Maintenance Analytics Digital Andon Digital Twin

OPTIMUS fleet health dashboard — Fleet overview — health, availability, active alarms, critical assets, open work orders.

OPTIMUS condition monitoring trends — Condition monitoring — vibration, temperature and current trended per asset.

OPTIMUS CMMS work orders — CMMS — work orders by type, priority and status, with schedule and spare parts.

OPTIMUS maintenance KPI and analytics — Maintenance analytics — MTTR, MTBF, SLA and work-order mix, per area and technician.

OPTIMUS plant digital twin — Plant digital twin — live asset status pinned onto the 3D model.

OPTIMUS digital andon operator panel — Digital andon — a categorised, one-tap channel from the floor to the people who can act, timestamped so response time becomes measurable.

My role: Platform architecture, sensor integration, analytics rules, CMMS workflow
Stack: TypeScript · React · Node.js · Python · 3D viewer
Sensing: Vibration · temperature · current · anomaly detection

Mobile · Flutter

TNS Energy Mobile — Android & iOS

The energy dashboard lives on a control-room PC. The person who can approve stopping the waste is out on the floor. This app puts today's consumption, its cost in rupiah, peak load against the contracted limit, and the open findings into their pocket. Tap a number and you get the arithmetic, the baseline and the curve it came from. Findings inferred from power alone are marked MEDIUM confidence with the reason printed underneath, because the product cannot prove more than that.

Flutter Dart iOS 12+ Android 8+ One permission: INTERNET

Three screens of the TNS Energy mobile app — Daily summary, consumption by area, and the open waste findings list. It would have been easier to show a big confident number — but this app tells a manager to spend money and interrupt a production line, so it has to show why.

My role: Product definition, information design, release
Platforms: iOS 12 and above · Android 8.0 and above
Permissions: INTERNET only — no camera, location, storage or contacts

Hardware · Firmware

Industrial IoT Node & sensors

Software people tend to assume the data is already there. On an Indonesian factory floor it frequently is not: the machine is twenty years old, it has no network port, and the only thing it will tell you is that it is on. The node is what we bolt on to change that — and it is the reason the platforms above have anything to read.

STM32 KiCad Encoder input Accelerometer Modbus MQTT

Machvory DIN-rail industrial IoT gateway, front and rear — The DIN-rail gateway that sits between the machines and the platform — RS-485 and Ethernet in, MQTT out. Product render.

My role: Circuit design, BOM, firmware, field commissioning
Core: STM32 (H7) with dual high-speed HTL encoder inputs
Installed: PT Javis Teknologi Albarokah · PT Astra Honda Motor (sample)

Web · Brand

machvory.co.id and the film behind it

I designed and built the company's public platform and the CMS behind it. Next.js 14 App Router, Prisma over MySQL, role-based admin access for four roles, first-party visitor analytics, and a lead pipeline that keeps attribution intact across a page reload. The whole site's information architecture lives in a single typed file, so the menu can never drift from the pages it points at.

Next.js 14 Prisma MySQL RBAC Docker nginx Blender

machvory.co.id homepage — The homepage above the fold.

Frame from the Machvory hero film — Frame 320 of a 24-second loop I built in Blender over Machvory's own production-hall model. The panels are cropped screenshots of the running products, not invented HUD graphics. Prerendered rather than WebGL — the earlier live scene cost load time, battery, and stability on mobile.

Why I am applying

Five years of making industrial data readable, and nearly all of it landed on a desktop dashboard. The one mobile product I have shipped reaches iOS through Flutter — which works, but means I have never actually written for the platform. Swift, SwiftUI, the Human Interface Guidelines, and the design side of building something people choose to open: that is the gap.

What I would bring is a domain most applicants do not have. I know what a maintenance technician does at 6am, why a plant manager distrusts a number, and what happens to an interface when someone is wearing gloves. I have also spent three years running a small engineering team, which mostly teaches you how to be wrong in front of people quickly enough to fix it.

I am not applying to change field. I am applying because the systems I build deserve a better front door than the one I currently know how to make.

I make factory data legible

Sensing & firmware

Industrial connectivity

Data & backend

Interface

Analytics & AI

Integration & delivery

Every number shows its work

No claim the product cannot prove

What I have built

TNS — Operational Performance & Energy Intelligence

OPTIMUS — Predictive Maintenance & CMMS

TNS Energy Mobile — Android & iOS

Industrial IoT Node & sensors

machvory.co.id and the film behind it

The stack I actually work in

Embedded & hardware

Industrial protocols

Mobile

Software & data

AI, 3D & delivery

Learning next

Where this came from

Director & Chief Technology Officer

Senior Researcher

Universitas Mercu Buana Yogyakarta

Why I am applying

Get in touch