Everguard.ai
A heavy-industry safety platform fusing computer vision, wearables, and IoT sensing
Everguard.ai takes a route different from other pure-vision workplace-safety AI: they feature "sensor fusion," integrating signals from computer vision, wearables, indoor positioning (RTLS), and various IoT sensors to build a more complete on-site risk picture than relying on cameras alone. The flagship platform is called Sentri360.
Features and application scenarios
Why go to the extra trouble of adding wearables? Because heavy-industry sites have too many places cameras can't see — a steel plant's furnace area, high-temperature dust, behind obstructions. Pure-vision solutions simply fail in these environments. Everguard's approach is to make the devices on workers' bodies also become sensing nodes, working with the positioning system to know where each person is, then cross-comparing with image data. They call this concept "Worker-Centric AI," emphasizing that it's not just a dashboard for managers — workers themselves also receive real-time danger alerts. Real-world applications concentrate on high-risk heavy industries like steel and metal processing. Note that their official site has Vercel's bot-protection mechanism, so accessing it in a way other than a normal browser may get blocked.
Suited to high-risk sites where "cameras can't see everything" like steel, metal smelting, and heavy manufacturing. Taiwan's steel and metal-processing industry is actually a very fitting customer base — the China Steel system, and casting and heat-treatment plants nationwide, whose high heat, dust, and blocked sightlines are the fatal weaknesses of pure-vision solutions. But conversely, the adoption cost is also high: you don't just buy software but also need to equip every worker with devices and build positioning infrastructure — a whole engineering effort.
Main features
- Sentri360 platform: integrates computer vision, wearables, RTLS positioning, and IoT sensing
- Sensor-fusion analysis, filling the blind spots of a single image source
- Worker-side real-time danger alerts (not just a management dashboard)
- Edge-computing processing to lower latency
- Designed for heavy-industry scenarios like steel and metal processing
- Real-time personnel-location tracking and danger-zone proximity alerts
Common uses
- Alerts for personnel approaching danger zones in a steel plant's furnace area
- Detection of personnel intrusion within a heavy machine's operating range
- Monitoring workers' physiological state and location in high-heat, dusty environments
- Real-time personnel headcount and location during emergency evacuation
- Root-cause analysis of heavy-industry workplace-safety events
Key Features
- Sentri360 platform: integrates computer vision, wearables, RTLS positioning, and IoT sensing
- Sensor-fusion analysis, filling the blind spots of a single image source
- Worker-side real-time danger alerts (not just a management dashboard)
- Edge-computing processing to lower latency
- Designed for heavy-industry scenarios like steel and metal processing
- Real-time personnel-location tracking and danger-zone proximity alerts
Pros
- Multiple sensing sources are far more reliable than pure vision in high-heat, dusty, obstructed heavy-industry environments
- Alerts go directly to the workers, not just a report for managers after the fact
- Higher specialization for heavy industry than general-purpose safety platforms
- Location data has real value for emergency response when an accident occurs
Cons
- Adoption cost clearly higher than pure-image solutions, involving wearable procurement and positioning infrastructure
- Wearable compliance is an old problem — if workers don't wear them, the system is half-blind
- The official site has bot protection, so getting public info is less easy than for peers
- Pricing not public; the privacy dispute of personnel-location tracking is more sensitive than pure imagery
Use Cases
- Alerts for personnel approaching danger zones in a steel plant's furnace area
- Detection of personnel intrusion within a heavy machine's operating range
- Monitoring workers' physiological state and location in high-heat, dusty environments
- Real-time personnel headcount and location during emergency evacuation
- Root-cause analysis of heavy-industry workplace-safety events
Editor's Note
Editor's note: I quite appreciate Everguard putting 'Worker-Centric' on the table. Most workplace-safety AI is, frankly, made for management — dashboards, compliance reports, audit numbers — while the workers themselves get nothing from it. They at least send alerts to the workers, giving the system direct value to the people on site. This isn't just product design but the key to adoption success: a system that makes workers feel 'this is protecting me' will have a completely different compliance rate.
FAQ
Why are wearables needed — aren't cameras alone enough?
In an environment like a steel plant, really not. Thermal turbulence from high heat, dust, and lots of blocking structures greatly lower image-analysis reliability. Wearables plus a positioning system can keep providing data where cameras can't see — this is Everguard's core design reason.
What if workers don't want to wear them?
This is the most realistic problem for this kind of solution, and there's no technical fix. In practice it relies on integrating the device into gear workers already have to wear (helmets, vests), making it valuable to the workers themselves (e.g., a case where a real-time alert saved a life), and clear data-use boundaries. Relying purely on mandates usually doesn't last.
How do I choose between it and pure-vision Intenseye and Voxel?
It depends on your site environment. For a general factory, warehouse, or logistics center with good camera views, a pure-vision solution is cheap and enough; for a steel plant, casting, high-heat high-dust, heavily obstructed heavy industry, the necessity of sensor fusion emerges. Don't pay extra for capabilities you won't use.
Related AI Tools
MOVUS
Australia's AI machine health monitoring that goes beyond alerts to tell you what to fix first
Luxonis
OAK series edge AI depth cameras that run computer vision directly on the device
Athena Security
Proactive concealed weapon detection using computer vision and walkthrough sensors—alerting security before a weapon is even drawn.
SoundThinking
An Operational AI Platform Unifying Gunshot Detection, License Plate Recognition, and Weapon Screening for Public Safety
Mark43
Cloud-Native Public Safety Platform for Law Enforcement and Dispatch with Built-in AI Reporting and Shift Briefings
Waste Robotics
AI-powered waste sorting robots utilizing hyperspectral imaging to detect material differences invisible to the naked eye.