Skip to content
Home: Anjaneya AutomationFrom Idea to Intelligent Machines
Industrial automationStatus: Working prototypeAI inside

Shree-Mazu Workbench: AI-assisted industrial engineering platform

Configure devices visually, define automation logic, generate embedded code and deploy to custom control hardware. A local AI assistant turns plain-English requests into rules.

Illustration of the Workbench flow: devices configured and mapped on screen, code written with the AI assistant, then built and deployed to a controller board that drives a stepper motor
Screenshot of the Shree-Mazu Workbench running a rule in its simulator, with a push button, two stepper motors, the event log and the speed control
Illustration of the Workbench: devices wired on the canvas, and the AI assistant turning a request into a rule

The challenge

Small automation jobs still mean ladder logic or embedded C, so every change waits for a specialist. We wanted a machine builder to set up the devices, write the logic and put it on real control hardware without writing firmware.

What we built

The Shree-Mazu Workbench is a desktop platform for the whole job. Place sensors, motors and outputs on a canvas and map each one to a pin on the controller. Describe the behaviour to a local AI assistant, or build it from visual logic blocks, and test it in simulation. The Workbench then generates the embedded code, builds it and flashes it to our SMZ-F4000 controller in one step, and a live monitor shows I/O values, events and logs while the machine runs.

Engineering highlights

  • A device library and canvas: sensors, motors and outputs mapped to the controller's pins with a click
  • An AI assistant on a local engine turns plain-English requests into rules, with no cloud account needed
  • Google Blockly blocks for motion programs, and for logic beyond simple rules
  • Simulation mode to test the logic before the hardware is wired
  • Embedded C generated, built and flashed to the board in one step
  • A live monitor showing I/O values, events and logs

From canvas to running machine

Set up the devices, define the logic and test it, then generate the firmware and put it on the board, all in one application.

The Workbench at work, sped up 2×: two stepper motors are added and wired to the board, a motion program is built from blocks, a rule runs it when a button is pressed, and the simulator shows it running.
  1. Configure devices

    Pick sensors, motors and outputs from the device library and map each one to a pin on the controller.

  2. Define the logic

    Describe the behaviour to the local AI assistant, or build it from Blockly blocks, then test it in simulation.

  3. Generate the code

    The Workbench writes the embedded C from the device map and the rules, and builds it.

  4. Deploy and monitor

    Flash the firmware in one click, then watch live I/O values, events and logs.

From Idea to Intelligent Machines

Have an idea, a product or a machine to improve?

Tell us what you want to build or the machine you want to modernize. We reply within 24 hours on working days. We sign an NDA before you share details.