alwaysAI puts trained AI models at the point of operation, on the excavator, at the dock door, above the tray line. Eyes on the edge means improvement in the moment: intelligence that arrives while the work is still happening, not after the shift ends.
Eyes on every cycle, every shift.
DigSight, HaulSight, DrillSight, and BeltSight, computer vision at the point of operation across surface and underground. Every major asset, continuously measured. Improvement in the moment, results in weeks.
Eyes on every movement.
LoadSight puts computer vision at the dock door, automated pallet verification, dispatch monitoring, and safety compliance in real time. Reducing manual scanning and catching errors before the trailer pulls away.
Eyes on every tray.
FoodSight puts computer vision above the tray line and prep station, verifying every meal against the dietary order, flagging errors before service, and building a complete audit trail across every shift.
alwaysAI puts trained AI models at the point of operation, on the dig unit, above the belt, at the go-line. For mining operations, that means continuous measurement of the factors that drive overall equipment effectiveness: availability, performance, and quality. Eyes on the edge. Improvement in the moment.
alwaysAI's computer vision runs on-site, processing camera footage from dig units, haul trucks, drills and fixed plant in real time. Every cycle measured. Every event logged. Every shift building a ground truth record of how your operation actually performs.
Deployed across surface, plant and underground operations.
PPE, hazardous zones, perimeter, fire & spill, available as add-on modules across all products.
Eyes on the edge is not a monitoring play. It is a performance play. Every mine has an OEE gap, the difference between what an asset produces and what it is capable of. When every swing cycle is measured, every haul truck timed, and every GET tooth watched continuously, that gap becomes visible. Availability losses, performance losses, quality losses: the cameras see them all.
The mining companies who get the most from alwaysAI are not running computer vision as a pilot program. They are embedding it as a permanent measurement layer, the ground truth that sits beneath every production review, every safety conversation, and every coaching session. Improvement in the moment. Improvement in the moment.
Continuous PPE, exclusion zone, and proximity monitoring, every shift, every access point, documented.
GET loss, oversize rock, and foreign objects detected at the point of operation. Material quality losses caught before they cause downstream stoppages or crusher damage.
Cycle times, fill factors, swing rates, idle frequency and active operating time as a percentage of shift, the performance and availability components of OEE, measured continuously at the asset.
Every cycle, event, and alert logged with timestamp and image, ground truth for shift handovers, investigations, and improvement programs.
LoadSight puts computer vision at the dock door and warehouse floor. Automated pallet scanning and verification at speed, dock scheduling visibility, forklift safety monitoring, and warehouse automation data, all from cameras already in your operation. Errors caught before the trailer pulls away.
alwaysAI's LoadSight suite brings computer vision to the pace and scale of modern logistics operations.
Integrating machine vision at outbound dock doors automates pallet scanning and verification, eliminating the need for manual checkers. Warehouse automation starts at the point where freight moves: dock scheduling becomes measurable, forklift safety becomes continuous, and pallet tracking becomes a data stream rather than a manual count. alwaysAI LoadSight cuts outbound shipping errors to virtually zero, reduces labour requirements at the dock, and accelerates loading efficiency by up to 20% in peak season by removing manual bottlenecks.
alwaysAI puts computer vision at the point of service, from the tray line to the stadium concourse. FacilitySight watches your venue. TraySight verifies every meal. Eyes on the edge. Improvement in the moment.
Computer vision at the point of service, from the stadium concourse to the hospital tray line.
FacilitySight puts computer vision across your venue, monitoring concourse flow, queue build-up, staff deployment, and safety compliance continuously across every zone and every shift.
TraySight verifies every assembled tray against the dietary order before it leaves the kitchen, flagging errors in real time and building a complete timestamped audit trail across every meal service.
Deployed in production across underground and open pit operations in North America, Africa, and Australia, and in logistics and food service operations globally. From first call to live production in weeks.
A Formula 1 tire change takes 2.2 seconds. That speed is only possible because the workshop behind it spent years designing the wheel nut, training the crew, and laying out every tool in exactly the right place. The fast part is only fast because of the preparation. That is how alwaysAI works.
We built the platform, the model training infrastructure, the edge deployment framework, the integration layer, the failure mode library, over eight years of production deployments in some of the most demanding industrial environments in the world. When we deploy on your site, you are not the first. The preparation was done long before we arrived.
"alwaysAI makes sites visible and measurable. Powered by AI and computer vision, our tools transform camera footage into clear, objective “ground truth” insights about how work is actually performed across a wide range of industrial operations."- Steven Longley, CEO, alwaysAI
Great results depend on good quality video footage and connections. We work with your team and partners to ensure cameras and compute are set up, models are trained, and data can flow from our application to the right systems.
From first call to live production in weeks, not months.
We assess your existing camera infrastructure, connectivity, and the operational processes you want to measure.
AI models are trained on footage from your specific site, your equipment, your environment, your conditions.
Hardware and software deployed on site, integrating with existing SCADA, historians, and reporting systems.
The system goes live. Models improve with every shift as they learn more about your operation.
The latest thinking on computer vision, industrial AI, and operational intelligence, from the alwaysAI team and the industries we serve.
alwaysAI is growing fast. We are deploying computer vision across some of the world's most demanding industrial environments and we are looking for people who want to build things that matter.
Our software runs on mine sites, in warehouses and in commercial kitchens. When something we build works, you can see the result immediately, in tonnes moved, injuries avoided, or waste reduced. There is no ambiguity about whether the work matters.
We are a lean, experienced team with a flat structure. Everyone contributes to decisions, everyone talks to customers, and everyone understands the product end to end. If you want to own your work and see it ship quickly, this is the right environment.
We have found genuine product-market fit in a hard space. The technology works, the customers are renewing, and the pipeline is growing. Joining now means growing with us from a position of momentum rather than uncertainty.
The team spans multiple time zones and we are comfortable working asynchronously. We care about output and ownership, not hours at a desk. We get together when it counts and trust each other in between.
All positions are open to remote candidates with the right experience.
Build and optimize CV models for deployment on edge hardware in demanding industrial environments. Work closely with customers to tune models to their specific site conditions.
Design and maintain the infrastructure that gets our models running reliably on NVIDIA edge hardware in remote mining and logistics environments.
Own the full sales cycle with Tier 1 mining and resources companies. Navigate complex procurement environments and articulate a compelling technical and commercial case to operational and executive audiences.
Drive adoption and value realisation across our deployed customer base. Work directly with site teams to ensure our products are delivering measurable outcomes.
Build the reporting, dashboards and integration layer that connects our CV platform to customer systems including SCADA, historians and production reporting tools.
We are always interested in hearing from strong engineers, sales professionals and operational specialists. If you think you can contribute, send us a note and tell us what you are good at.
Tell us about yourself and attach your CV. We review every application personally and aim to respond within five business days.
Thanks for applying. We review every application personally and will be in touch within five business days.
Request a pre-configured camera kit, capture a short window of footage from your own site, and send it back. We run our computer vision models against your real assets and return a report showing exactly what we detect, at no cost and with no commitment.
Everything ships ready to point at your asset. Capture around 30 minutes of footage, then mail the kit back using the included guide.
Send us video footage from a camera onsite and we will show you exactly what our models detect in it.
Not sure whether you have eyes on the right things? We can send you a GoPro in a ruggedized Pelican case, a two-page setup guide, and a return pack. Capture 30 minutes of footage from an asset you want to improve. Send it to us. We will run our models against it and show you exactly what we can detect, before you commit to anything.
An assessment de-risks the conversation for both sides. You see what is detectable on your site before committing to a deployment. We understand your equipment, environment, and data quality before scoping a solution. The result is a faster, more accurate deployment when you proceed.
No stock demos or simulated data. The report is built from your equipment, your site conditions, and your operating environment.
The kit, the analysis, and the report are provided free of charge. There is no obligation to proceed to a paid deployment afterwards.
Once we receive the kit back, your report is typically ready within 15 business days, covering detections, edge cases, and deployment recommendations.
Tell us about your site and the asset you want assessed. We ship a pre-configured camera and rugged case to your address.
Follow the included guide to position the camera and record around 30 minutes of normal operation.
Pack the kit back into its case and send it back to alwaysAI.
We run our models against your footage and return a report showing detections, accuracy, and recommended next steps.
Fill in your details and we will confirm your request with a call then ship the kit within 48 hours of confirmation. Tell us which asset type you want to assess and we will make sure the kit is configured appropriately.
Thanks - a member of the alwaysAI team will be in touch to confirm shipping details before your kit goes out.
Every mature mining operation maintains a list of fatal risks. BHP, Anglo American, Rio Tinto, and Glencore all converge on the same recurring categories, hazards that have killed people in the industry and that continue to appear at the top of fatality data year after year.
This paper goes through each of the 10, explains why it kills, and details how computer vision can detect the hazardous condition before it becomes fatal, using cameras already pointed at your major assets.
This whitepaper draws on the published fatal risk frameworks from BHP, Anglo American, Rio Tinto, and Glencore, cross-referenced against ICMM annual fatality data for 2023 to 2025 and guidance from the NSW Resources Regulator.
The 10 risks presented are the categories that appear most consistently across all of these frameworks and account for the majority of mining fatalities each year. They are not theoretical. Every one represents circumstances where people in the industry have died.
For each risk, the paper identifies the specific camera-observable precursors that typically precede a fatal event, and explains how computer vision, deployed on existing site infrastructure, can detect those precursors in real time before they become fatal incidents.
Sources: BHP Fatal Hazard Protocols · Anglo American Fatal Risk Standards · Rio Tinto Critical Risk Management Framework · Glencore Fatal Hazard Protocols · ICMM Safety Performance Reports 2023-2025 · NSW Resources Regulator
Enter your details and we will send the whitepaper to your email immediately. We will also follow up once to ask if you have questions, nothing more.
The whitepaper is on its way. If you have any questions, reply to that email and one of our mining specialists will respond.