NexOps
Local-first MES and SCADA platform for shop-floor operations
NexOps provides a local operational workspace uniting machine telemetry, production schedules, alarms, and shift OEE. As AI, Simulation & Platform Lead, I design repeatable factory simulations and lead the software runtime that streams live telemetry into operator dashboards.
Tech stack
Frontend
- React
- TypeScript
- Vite
- Ant Design
- Apache ECharts
- Zustand
- React Router
- PixiJS
Backend & API
- FastAPI
- Python
- SQLAlchemy
- Alembic
- Pydantic
- JWT
- HTTP/REST
- SSE
Data & knowledge
- TimescaleDB / PostgreSQL
- Redis
- EMQX
- MQTT
- MQTT.js
- Celery
Edge & simulation
- aiomqtt
- Node.js
- PLC simulator
- Bambu Lab LAN
Testing
- Pytest
- Vitest
- Playwright
Platform & observability
- Docker Compose
This video could not be loaded. Open video file
The problem
On many factory floors, machine telemetry, production schedules, alarms, and downtime logs are scattered across separate screens, spreadsheets, and paper notes. NexOps brings these into a single local MES/SCADA interface for operators, maintenance engineers, and plant managers.
What it covers
The prototype brings together live machine telemetry, work-order scheduling, alarm routing, and shift OEE. Deterministic simulation scenarios let us replay standardized shifts to verify telemetry ingestion, alarms, and OEE calculations.
How it works
- 01Device or deterministic simulator
- 02MQTT messaging through EMQX
- 03FastAPI services and TimescaleDB
- 04Dashboards, alarms and OEE
My role
NexOps is an active working prototype built for investor demonstrations and technical evaluation. As AI, Simulation & Platform Lead, I lead the software architecture, simulation engine, and integration adapters, while coordinating MQTT/EMQX messaging with our IoT hardware teammate.
- Designed and implemented deterministic machine and shift simulations to verify production schedules, alarms, and OEE calculations.
- Built the backend data pipeline ingesting MQTT/EMQX events through FastAPI into TimescaleDB and streaming live state to dashboards.
- Developed the local network adapter for a Bambu Lab 3D printer, demonstrating local device control in a recorded trial.
Working prototype
Telemetry from devices and simulators streams over MQTT through EMQX. FastAPI saves time-series data into TimescaleDB, while React, Apache ECharts, and PixiJS render shift schedules, andon boards, alarms, and OEE reports. The Bambu Lab integration shows local hardware control working alongside the simulation runtime.
