💡 "Life is your real journey, while work is just the oil that keeps it running smoothly."
I'm building industrial software systems and AI-native tools for manufacturing environments,
with a strong focus on Python, FastAPI, WMS / WCS / PMMS, Industrial IoT, and human-guided AI.
I enjoy designing practical architectures, cleaning up messy workflows, and helping machines talk to software 🤖🏭.
- 🔭 Currently exploring: AIIS — AI-Native Industrial System
- 🏭 Working around: WMS / WCS / PMMS / MES-L2 / Digital Twin / Industrial IoT
- 🧩 I like turning messy factory workflows into clear APIs, task states, audits, and operator-friendly tools.
- 🛡️ I care about safe automation: AI can diagnose and propose, but risky actions need approval, rollback, and device-side boundaries.
- ⚡ Fun fact: I still think industrial automation systems should feel as smooth as good music 🎶
AIIS is my long-term exploration of industrial systems that can perceive, remember, coordinate, recover, and evolve under human governance.
It is not just a chatbot or a single agent. I use it as a place to organize ideas from real industrial software projects:
ERP · MES / L2 · WMS · WCS · SCADA · PLC · Digital Twin · Industrial IoT · Data Governance · AI Agents
Systems should carry complexity so humans do not have to.
| Area | Project / Track | What it explores |
|---|---|---|
| 🧠 Industrial AI architecture | AIIS | Runtime architecture, agent runtime, workflow, governance, rollback, skills, cases, and demos |
| 📦 Production material management | AIIS-PMMS | Laser-cutting remaining-material backend, SQL Server, OpenAPI contract, Template.xlsx export, production report records |
| 🏬 Warehouse runtime | AIIS-WMS | Inventory, orders, AGV task monitoring, multi-client runtime, hotspot API optimization |
| 🚦 Warehouse control | AIIS-WCS | Command intake, task state, resource locks, worker execution, device-control boundaries |
| 🌐 Digital twin | AIIS-DataTwin | Wis3D-compatible sandbox, GLB/model tooling, simulation worker |
| 🏭 Factory runtime | AIIS-PlantOps | L2 production runtime, Vue/FastAPI/Celery/Tauri direction |
| 🛠️ Engineering file workflow | FastBOM | Manufacturing engineering-file workflows, desktop operator tooling, SolidWorks/DXF boundary |
Some factory-derived repositories are private or published as sanitized / documentation-oriented snapshots 🏭🔒.
I keep public repositories focused on architecture, reusable patterns, demos, and project learning.
Backend & API
Python · FastAPI · Flask · SQLAlchemy · Alembic · Pydantic · JWT / RBAC · OpenAPI
Data & Runtime
SQL Server · MySQL · PostgreSQL · SQLite · Redis · Celery · task state machines · audit logs
Industrial / IoT
WMS · WCS · PMMS · MES / L2 · PLC boundary design · MQTT · Socket protocols · AGV / task scheduling
AI & Agent Systems
AI application architecture · agent runtime · RAG-ready knowledge · sandbox validation · human approval gates
Frontend / Client Surfaces
Vue · Tauri · Qt / QML · PySide6 · WPF · Streamlit · NiceGUI
DevOps / Tools
Docker · Docker Compose · Linux · GitHub Actions · uv · PyInstaller
I do not think industrial AI should mean “let an LLM directly control production.”
The direction I care about is bounded autonomy:
observe -> understand -> propose -> verify -> classify risk -> request approval -> execute within boundary -> record evidence
AI can help with diagnosis, testing, documentation, optimization, and controlled maintenance workflows.
High-risk production or device-side actions should stay approval-gated, auditable, reversible, and protected by local control and hardware safety boundaries.
- 💼 LinkedIn: Jason Bu
- 🔗 GitHub: @jasonbu163
- 👾 Gitee: @jasonbu163
- 📫 Email: jasonbu163@163.com
✨ Thanks for visiting my profile! Always open to conversations around industrial software, manufacturing IT, AI-native systems, and human-guided automation 🚀