Every Weld Graded in 38 ms, On the Line, Without a Network
Inline Weld Seam Inspection Cell
38 ms
inference latency, inside the 45 ms index window
100%
of welds inspected, up from 2.5% sampling
Melbourne · Engineering studio
Engineering that ships to the factory floor.
We build edge AI, embedded systems, IoT fleets and industrial automation - and the software that runs them. One team from schematic to production line.
15
production programs shipped
8
industries with systems in service
4
working demos you can open right now
2022
studio founded, Melbourne
Who we are
One team, from atoms to dashboards.
Most projects that mix hardware and software fail at the seam between them. We hold both sides, so there is no seam to fail at - and no vendor to point at when it does.
Constraint first
We start from the budget that actually binds - latency, power, thermal, certification date - and design backwards from it. Most failed projects picked the architecture first and met the constraint never.
Production is the target
A demo that works on a bench is not a result. We take designs through DFM, pilot runs and commissioning, because the hard problems surface at unit 100, not unit one.
Documented for the next team
You get schematics, firmware, test procedures and a handover pack a team that does not exist yet could pick up cold. Your IP, transferred on payment.
Capabilities
Ten practice areas, one accountable team.
Take a slice or hand us the whole program. Either way one team owns the result and one person answers the phone.
Edge AI Development
Inference on constrained hardware against real latency and power budgets - quantised models, measured, not benchmarked in a notebook.
Embedded Systems
Boards, firmware and integration - schematic to production line, by one team with the whole product in view.
IoT Development
Device fleets that stay online - LoRaWAN, cellular and mesh, with provisioning, OTA and telemetry that survives bad links.
Industrial Automation
PLC, motion and OT/IT integration on plants that cannot stop. Retrofit paths for machines older than the engineers running them.
Computer Vision
Optics, illumination and models treated as one system - because most vision projects fail on lighting, not on the network.
Robotics
Autonomy stacks from perception to actuation, with the safety interlocks and manual-override paths that make them deployable.
Cloud Software
Ingest, time-series storage and APIs sized for device fleets, with the offline-first behaviour edge deployments actually need.
Hardware Product Development
End-to-end product engineering - industrial design, mechanical, electronics, firmware and manufacturing handover under one roof.
Selected work
Built, shipped, running in production.
A cross-section across edge AI, industrial IoT, robotics and enterprise platforms. Every entry carries its constraints, what we built, and what it measured afterwards.
Partial-Discharge Sensing with Evidence-Rich Alerts
13kV RMU Fault Detection System
Early
fault detection before outage
Evidence-rich
alert payload
Shift Performance, Machine Utilisation and Quality in One Pane
Manufacturing Execution Dashboard
61% → 74%
OEE across 34 machines in eight months
34
machines unified across four vendors and two decades
Ground-up Autonomous Platform for Commercial Grounds
Autonomous Zero-Turn Lawn Mower System
360°
safety perception coverage
RTK-grade
positioning accuracy
Electricity, Water and Environment Across a 40-Site Portfolio
Building Energy & Utility Monitoring
12.4%
weather-normalised electricity reduction, year one
40
sites unified across BMS, Modbus and LoRaWAN
Multi-Tenant SaaS With Role-Optimized Workflow
Fleet Management IoT Platform
Multi-tenant
with pluggable integrations
Role-optimized
dispatcher / manager / driver views
Live demos
Don’t take our word for it. Open the build.
4 working interfaces from systems we design, running in your browser on simulated data. No login, nothing to install.
Live
AI Ponics Autonomous Hydroponics Farm
6,500 Plants Under Continuous Autonomous Control
Live
13kV RMU Fault Detection System
Partial-Discharge Sensing with Evidence-Rich Alerts
Live
Fleet Management IoT Platform
Multi-Tenant SaaS With Role-Optimized Workflow
Industries
Deep context, not generic consulting.
Eight sectors where we have production systems in service - and know which constraints actually bite.
AgriTech
Controlled-environment agriculture, autonomous farms, and the hardware, firmware, and analytics beneath them.
Robotics
Outdoor autonomous platforms, retrofit kits, perception stacks, and fleet coordination.
Industrial IoT
Medium- and high-voltage monitoring, predictive maintenance, and secure industrial telemetry.
Automotive
Driver assistance, telematics, fleet management, and connected-vehicle security.
Smart Industry
OEE dashboards, production line controllers, digital twins, and industrial automation.
Energy
Energy monitoring, solar performance, battery storage, smart metering, and microgrids.
Smart Cities
Street lighting, air-quality networks, water monitoring, waste routing, and shared city IoT platforms.
Software & SaaS
Device management, time-series analytics, automation rule engines, and multi-tenant cloud.
Technologies
The stack, stated plainly.
106 technologies across 15 shipped programs - chosen per constraint, not per preference. If a project needs something not on this list, we say so.
Edge AILoRaWANMQTTESP32PLCComputer VisionTime-Series DBReactSTM32NB-IoTCFD airflow analysisSLA prototypingDifferential driveLiDARGNSS RTKROS 2CAN busBLDC motor controlDrive-by-wireHMIshared control algorithmsInductive/contact chargingvisual fiducialsPartial-discharge sensingcurrent/voltage meteringedge AITime-series DBstreaming analyticsanomaly detectionDSPML classifierswaveform databasesIEC 61850 awarenessedge gatewaycloud telemetryMulti-tenant cloudgeofencingworkflow automationGNSScellular modemSecure gatewaysigned commandsmTLScellularshared platformMesh networkingluminaire controllersCMS
How we work
A process that survives contact with reality.
Four phases with a real deliverable at each gate. You always know what you paid for and what ships next.
PHASE 01
Discover
A paid week that ends in a written architecture, a budget range, and the risks named. You own the output whether or not we build it.
PHASE 02
Design
Schematics, mechanical, and software architecture in parallel - with the long-lead and certification items pulled forward first.
PHASE 03
Build
Iterative hardware and firmware revisions against real bench and field testing. You see every revision, not just the last one.
PHASE 04
Deploy
Pilot, commission and hand over - with the documentation pack a second team could pick up cold.
Testimonials
What clients say.
They built the device and the software behind it.
We came to Incendio with a rough concept and not much else. They built the device and the software that runs it, then stayed hands-on until it held up in the field, not just on the bench. What surprised me was how much they flagged early instead of at the end.
Asif
CEO, VEIVARSLER
One team instead of three vendors.
The best part was not having to manage three separate vendors. One team owned the electronics, the firmware, and the dashboard, so when we hit a snag it got sorted in days. They were honest about the risky parts up front, which we appreciated.
Operations Director
Sautronics
Hardware and software that finally talk.
Most of the shops we spoke to did either hardware or software. Incendio did both and got them talking to each other. Our field devices finally feed the dashboard our operators actually use.
CTO
HSA Security Solutions
The discovery week earned its keep.
The paid discovery week was worth it on its own. We walked away with a real plan, budget ranges, and a written list of the risks. The build tracked to it and the invoices matched.
Product Lead
THAS MultiTech
Prototype fast, document properly.
They prototype fast but document like a much bigger company. By the time we reached our first production run, the handover pack had already answered most of our questions.
Engineering Manager
Hardee's
Leadership
The people accountable for your program.
You work with these three directly. There is no account manager between you and the engineers.
Shoaib Ahmad
Founder & CEO
Sets the engineering bar and runs client programs. Background in embedded systems and product delivery.
Hassan Ali
Co-Founder & CTO
Owns architecture across hardware and software - robotics, firmware, and cloud, the whole stack.
Dr Azad
Director & COO
Runs operations, partnerships, and the delivery pipeline. Makes sure what’s promised is what ships.
FAQ
Straight answers to common questions.
Programs where hardware and software have to work as one system - edge AI, embedded devices, IoT fleets, industrial automation, and the platforms that sit above them. We are a poor fit for pure marketing sites or standalone mobile apps with no device behind them.
Most programs start with a paid discovery week that produces a written architecture, a budget range, and a named risk list. After that we quote fixed-scope phases or a monthly retainer. You own the discovery output either way.
We take designs through DFM, pilot runs and production handover, and we work with contract manufacturers in Australia and overseas. We do not own a factory - we make sure the design survives one.
You do. All foreground IP created for your program transfers to you on payment. We retain only pre-existing tooling and libraries we bring in, licensed to you perpetually.
A working prototype usually lands 8-16 weeks after discovery, depending on whether custom silicon, certification or tooling is on the path. We pull long-lead items forward in week one so they are not what delays you.
For a first system, a five-person team across electronics, firmware, mechanical, cloud and ML takes most companies a year to hire. We are faster to start and stop, and we hand over documentation built for a team that does not exist yet. For steady-state operation of a shipped product, in-house wins - and we plan the handover for it.
Contact
Tell us what you’re building.
Send the constraint that worries you most - a latency budget, a power budget, a certification date. We reply within one business day, and we will say so if we are not the right team.
Direct
Studio
Melbourne, Victoria, Australia
Service area: Australia-wide
Hours
Mon-Fri, 9:00-18:00 AEST