Work
Selected work
Track and trace, SCADA, warehouse management, vision, and reporting work - described by industry and scope.
Client names are withheld by default. Ask if you need them.
-
MES and SCADA ownership for two battery manufacturing sites
Ignition MES and SCADA ownership for two battery manufacturing sites: track and trace, gate checks, database front-ends, and production reporting.
- Built a reusable library of Ignition templates for track and trace, gate checking, and database front-ends, so new lines and sites commission from a proven baseline
- Process requirements turned into MES workflows and operator screens, worked out directly with production and quality teams
Ownership
- Ignition Perspective
- MES
- SCADA
- Track and trace
- Production reporting
-
Ignition Perspective standards across three plants in startup phase
Ignition Perspective development across three plants in startup phase, setting the patterns and standards other plant engineers built on.
- Connected OPC-UA servers and Mitsubishi PLCs into plant-wide data collection for new production lines
Standards and integration
- Ignition Perspective
- OPC-UA
- Mitsubishi PLCs
- Data collection
-
Battery gigafactory SCADA
Ignition Perspective SCADA for a battery manufacturing gigafactory, from track and trace through MES administration.
- Track and trace and gate checking
- Recipe management
- MES admin dashboards and reporting
Client delivery
- Ignition Perspective
- Microsoft SQL Server
- Docker
- Git
- Raspberry Pi
-
Plant overview with live AGV tracking - defense industry
A plant overview showing the status of every station plus the real-time position and status of dozens of AGVs.
- Pan-and-zoom container for the whole plant
- Abstracted into a reusable template for fast deployment
Client delivery
- Ignition Perspective
- FactoryTalk
- AGV tracking
-
Warehouse management system and pick-to-light in Ignition Perspective
Warehouse management system in Ignition Perspective for a defense industry client, delivered with a small team, plus a pick-to-light system for the same warehouse.
- Warehouse material management and material transfer history
- Picklist submission and a pick-to-light system for warehouse operations
- User management and reporting
Application build
- Ignition Perspective
- PostgreSQL
- Git
- Pick-to-light
-
MES extensions: ASRS, conveyor tracking, and reject prevention
Extended an in-house Ignition MES with material handling, finished goods tracking, and defect-hold logic on a just-in-time production line.
- ASRS and gantry logic
- Finished goods conveyor tracking with Keyence scanners
- Andon logic that detects repeated defects and holds the machine until maintenance responds
- Automated production tracking and reporting, plus MES fault resolution on a live line
MES development
- Ignition
- Python
- Microsoft SQL Server
- Cognex DataMan
- Keyence
-
Automated visual inspection on a live production line
Photo eyes on a live production line triggered a Raspberry Pi to photograph each part, and a TensorFlow model classified it as good or bad.
- Result displayed in Ignition Perspective, and a bad part stopped the conveyor
- Operator confirm or override corrections fed back into retraining the classifier
Machine vision
- TensorFlow
- Raspberry Pi
- Computer vision
- Ignition Perspective
- Python
-
Plant trigger display
A plant display relaying the information the floor and customer-facing teams needed to see at a glance.
- Finished goods inventory, trailer location, and inventory at the customer
- Customer broadcasts and build queue
- Plant overview with downtime and production status, plus audible alarms for production-critical stoppages
Plant display
- Ignition
- Python
- Microsoft SQL Server
- GPS tracking
- VOIP
-
Scheduled report pipeline as a systemd timer
A daily report pipeline designed as a systemd user timer with a retry wrapper and environment-based configuration.
- journald logging for every run
- 26-assertion acceptance test suite that runs against disposable sandboxes
Automation
- Bash
- systemd timers
- Testing
- Automation
-
Self-hosted AI agent stack
Runs the Hermes agent framework as supervised services (gateway and dashboard) on Linux and WSL2, backed by local Ollama models.
- systemd units, restart policies, and journald logging
- Remote access through a Cloudflare Tunnel and a bearer-token-gated Caddy reverse proxy
Platform
- Hermes
- Ollama
- MCP
- systemd
- Cloudflare Tunnel
- Caddy
- Python
Want the detail behind any of these?
We can walk through the architecture, the tags, and the parts that were hard.