Safer shifts, higher recovery, fewer surprises
Mining Technology
Software and AI for the platinum, gold and chrome operations of the North West and beyond — built by a team based in Rustenburg, at the centre of the Bushveld Complex.
- Reduction in unplanned downtime
- 34%
- Safety observations captured
- 4,2×
- Reporting time saved weekly
- 22 hrs
Our position
Why we build for mining
We are headquartered in Rustenburg, surrounded by the largest platinum operations on earth. That proximity matters: our engineers have been underground, they understand why a system that assumes connectivity fails at 1,200 metres, and they know the difference between a Section 54 and a Section 55. Mining is our deepest domain, and we build for its actual constraints rather than a generic industrial template.
Regulations we design against
- Mine Health and Safety Act 29 of 1996
- Mineral and Petroleum Resources Development Act 28 of 2002
- Protection of Personal Information Act 4 of 2013 (POPIA)
- DMRE statutory reporting requirements
- ISO 45001 — Occupational Health and Safety
- Mining Charter III social and labour plan reporting
Challenges
What people in this sector actually raise
These are the problems that come up in a first meeting, with the cost of leaving each one unaddressed.
- 01R1.2m–R4m per unplanned stoppage day on a mid-tier operation
Unplanned equipment failure
A single unplanned mill or conveyor stoppage cascades through the entire value chain, and the cost is measured in shifts rather than hours.
- 0230–40% of a safety officer's week spent on transcription rather than prevention
Safety compliance and reporting burden
Mine Health and Safety Act obligations generate enormous volumes of inspection, incident and training records, most of it still captured on paper and re-keyed days later.
- 03Data captured hours or days late, with accuracy degrading as memory fades
No connectivity where the work happens
Underground and in remote pit areas there is no signal. Systems designed for the office fail at exactly the point where data is created.
- 04Reconciliation consuming days per month and eroding trust in every report
Fragmented operational data
The SCADA historian, the ERP, the fleet-management system and the laboratory information system each hold part of the picture and none of them agree.
- 051–3% recovery left on the table, compounding across every tonne processed
Grade and recovery variability
Plant recovery drifts for reasons that are visible in the data long before they are visible in the numbers management reviews.
What we build
Systems for mining operations
Predictive maintenance
Vibration, thermal and current-signature models that flag developing failures weeks before they stop production.
Computer-vision safety monitoring
Automated detection of PPE compliance, exclusion-zone breaches and proximity events, with privacy-preserving on-edge processing.
Offline-first inspection apps
Underground-capable capture for inspections, permits, pre-use checks and incidents, syncing the moment a device reaches signal.
Unified operational data platform
One reconciled model across SCADA, ERP, fleet and laboratory systems, with lineage back to the source.
Recovery optimisation models
Plant models that correlate feed characteristics with recovery and recommend setpoint adjustments in real time.
Automated statutory reporting
DMRE and Mine Health and Safety Act reporting generated from source data with a complete audit trail.
AI opportunities
Where AI genuinely pays in this sector
Ranked by how proven the approach is. We are explicit about what is experimental, because deploying an experiment as though it were proven is how AI programmes lose credibility.
Conveyor belt condition monitoring
provenVision models detecting rips, misalignment and material spillage from existing CCTV, without new instrumentation.
Typical return6–9 month paybackRock fragmentation analysis
provenAutomated size-distribution measurement from images at the tip, feeding blast-design optimisation.
Typical return3–7% reduction in downstream milling energyGeological model refinement
emergingMachine learning over historical drilling and production reconciliation to reduce grade-estimation error.
Typical return1–2% improvement in mine call factorAutonomous ventilation-on-demand
emergingAirflow modulated to actual occupancy and equipment activity rather than a fixed schedule.
Typical return15–25% reduction in ventilation energyNatural-language operations assistant
emergingA grounded assistant that answers questions across standards, procedures, incident history and shift reports.
Typical returnHours returned per supervisor per week
How we deliver
Services and products that apply here
Services
Questions
Mining — common questions
The questions we are asked most often by operators in this sector.
Can your systems run underground with no network?
Yes. Offline-first is the default architecture for every mining application we build. Data is captured, validated and queued locally, then synced with conflict resolution when the device reaches signal.Do you integrate with our existing SCADA historian?
Yes — PI System, Wonderware, Ignition and OPC UA sources are all standard integrations. We read from the historian rather than replacing it.How do you handle POPIA with camera-based safety monitoring?
Processing happens on the edge and only the event is retained, not the footage. Faces are blurred before any frame is stored, and retention periods are configured to your policy and enforced automatically.Are you familiar with the local operating environment?
We are based in Rustenburg, in the middle of the Western Limb. Our team has worked on platinum, gold and chrome operations across the North West and Limpopo.
Working in mining?
Tell us what you are trying to change. We will tell you honestly whether we can help and what it would take.