Protocol-agnostic edge runtime. Five capability pillars that deploy independently or together. Hardware and software designed to grow together. Operating across India and the Middle East. Deployed in defense and space-agency programs.
Industrial OT teams want one operational view across every controller on their floor. They want canonical signals at every downstream system. They want predictive maintenance that actually triggers on real condition signatures, not on calendar schedules. They want multi-site visibility without losing per-plant operability. And they want all of this without ripping out the SCADA, historian, MES, and CMMS infrastructure that already works.
What they usually get instead is a stack of per-vendor monitoring tools, per-protocol custom scripts that break on firmware updates, integration platforms that are opinionated about the operations team's existing systems, and "AI-powered" dashboards that don't survive contact with real production reality. The promise of "Industry 4.0" gets diluted by tooling that doesn't match how industrial operations actually work.
Elpis was built differently. One protocol-agnostic edge runtime that speaks every controller's native protocol. Canonical normalization at the edge so every downstream system reads the same vocabulary. Hardware that handles the cases controllers don't cover. An analytics platform that models the real industrial hierarchy. Modular per pillar — pick what fits your floor, scale on your terms. Beside the systems you already run, not replacing them.
Platform principles. Five operating commitments that shape every architectural and product decision Elpis makes:
EdgeConnect (the protocol-agnostic edge runtime, Windows today + Linux on the roadmap) + Edge Gateway (ruggedized appliance, dual identity — standalone today, canonical EdgeConnect appliance once Linux ships). Polls existing controllers in their native protocols (FOCAS2, MTConnect, Brother HTTP, Modbus TCP, OPC UA Client, S7). FANUC MT-LINKi REST integration on the roadmap. Canonical vocabulary at the edge.
See the deep-dive → /capabilities/connectivity-edgemDAQ — ruggedized acquisition hardware for direct sensor reads where no PLC exists, where the PLC is locked, or where adding a PLC layer would be more expensive than acquiring the signal directly. Greenfield installs, PLC-bypass retrofits, remote and unmanned sites.
See the deep-dive → /capabilities/data-acquisitionmTracker — utilization, location, and OEE telemetry on equipment that doesn't speak a controller protocol. Multi-site asset visibility for plant operators; OEM service-hours billing and warranty triggers for OEM machine builders. Customer-controlled routing for OEM-channel deployments.
See the deep-dive → /capabilities/asset-intelligenceVAS (Vibration Analyser System runs on mDAQ — bearing issues, imbalance, misalignment, looseness, and cracks on rotating machinery (pumps, motors, gearboxes, fans, compressors), conveyors, and structural components) + E-IDOS (standalone sensor-agnostic appliance for hydraulic and lubrication oil-health — particle contamination, water saturation, oil flow; ISO 4406 / NAS 1638).
See the deep-dive → /capabilities/condition-monitoringEREMOS V2 — multi-tenant analytics platform. Models the real industrial hierarchy (PLANT → AREA → LINE → EQUIPMENT → SUB_EQUIPMENT). Computes OEE via Segments against your OEE definition. Persistent alarms with incident workflows. Per-tenant isolation by design.
See the deep-dive → /capabilities/operational-intelligencePillars deploy independently. A site can start with Connectivity & Edge (consolidate mixed-vendor monitoring), add Operational Intelligence (OEE + alarms + incident workflows), then scale into Condition Monitoring (predictive maintenance) when the maintenance program is ready. No "all-or-nothing" forced bundling. Each pillar carries its own commercial conversation; combinations are negotiated based on the actual deployment shape.
EdgeConnect and EREMOS V2 run offline-first by default. Cloud connectivity is opt-in, not required. Plants on isolated OT VLANs install and run the platform the same way as plants with internet access. When cloud is part of the deployment, the customer controls the routing — which signals route where, on what schedule, with what scope. Hybrid edge-plus-cloud combinations are common; the customer decides the boundary.
OEM machine builders embed mTracker in shipped equipment for service-hours billing, warranty triggers, and remote diagnostics — under customer-controlled routing where the OEM sees only what their customer authorizes. System integrators partner on multi-site standardization projects. Integrator engagements scope per project, not per long-term resale contract.
Maintenance and AMC providers across India and the Middle East deliver predictive maintenance and condition monitoring on customer floors using Elpis hardware (VAS for vibration, E-IDOS for oil-health) plus EREMOS V2 incident workflows. The AMC channel is an existing deployment reality, not a future partner program — customer-controlled signal routing makes AMC engagements scoped, auditable, and reversible per asset and per contract.
The same platform supports both OEM and AMC business models without requiring separate stacks. OEM machine builders embedding telemetry in shipped equipment and AMC providers delivering service on customer floors run against the same EdgeConnect runtime, the same EREMOS V2 incident workflows, and the same customer-controlled routing model. The deployment shapes differ; the platform doesn't fragment.
Commercial scope is established through architecture review and deployment scoping. Detailed pricing follows approved scope — per engagement, based on pillar combination, deployment footprint, and plant count. SKU grids, per-tag pricing, and subscription detail are published once the customer commercial baseline is established.
Read the full operational trust posture → /security
Named customer stories arrive in Phase 3 with explicit named-customer sign-off. Defense and space-agency customer names remain off-the-record per the locked external-claim policy.
What's the commercial model?
Modular per pillar. Each pillar (Connectivity & Edge, Data Acquisition, Asset Intelligence, Condition Monitoring, Operational Intelligence) carries its own commercial conversation; combinations are negotiated based on actual deployment shape. Pricing scopes per engagement based on pillar combination, deployment footprint (software-only vs appliance, per-plant count), and plant geography. Architecture review and scoping happen first; pricing follows the scope.
How does engagement typically scope?
Architecture review → scoping → phased rollout. Architecture review starts with the controller mix, sensor inventory, existing-systems boundary (SCADA / historian / MES / CMMS coexistence), and the per-plant deployment shape preference. Scoping translates the architecture into a concrete pillar combination + plant count + timeline. Phased rollout starts bottom-up by asset criticality — instrument the highest-criticality assets first, build threshold + workflow discipline there before scaling out.
Do you support OEM and AMC partner engagements?
Yes. OEM machine builders embed mTracker for service-hours billing, warranty triggers, and remote diagnostics under customer-controlled routing. AMC providers deliver predictive maintenance and condition monitoring on customer floors using VAS + E-IDOS + EREMOS V2 incident workflows. Both engagement shapes use customer-controlled signal routing — the customer decides which signals route to which OEM or AMC for which assets. AMC is an existing buyer reality across India and the Middle East today, not a future channel program.
What's the deployment philosophy?
Beside, not replacing. EdgeConnect and EREMOS V2 sit beside your SCADA, historian, MES, and CMMS — they don't take over operator HMIs, control logic, alarm acknowledgment, work-order management, or scheduling. Per-plant identity, not per-fleet — each plant runs its own EdgeConnect runtime with a per-gateway UUID; multi-site visibility comes from EREMOS V2 aggregating across per-plant runtimes. Offline-first by default; cloud connectivity is opt-in.
What does this NOT do — where do our existing systems stay?
Your SCADA stays where it is (operator HMIs, control logic, alarm acknowledgment workflows). Your historian stays where it is (long-term archive of record). Your MES stays where it is (work-order management, scheduling, labor tracking). Your CMMS stays where it is (maintenance system of record). EdgeConnect publishes to MQTT and exposes signals via OPC UA Server (your existing SCADA can subscribe to either). EREMOS V2 exposes OEE rollups, alarms, and reports via REST API, MQTT, and webhook integrations. Your existing systems consume canonical signals instead of vendor-specific ones. Full integration patterns → /architecture.
Where do you operate today and where do you ship?
Operating across India and the Middle East. Plants, multi-site operators, OEM machine builders, and AMC providers across both regions. Deployed in defense and space-agency programs. Customer names remain anonymized per the locked external-claim policy. International relevance without overclaiming "global." Geography expansion happens per scoped engagement — when the customer mix and deployment scale justify it, not as a forced multi-region positioning claim.
Whether you're evaluating Elpis for a single plant, scoping a multi-site standardization, or scoping an OEM / AMC partner engagement — bring us the operational scope and the existing-systems boundary. Architecture review runs against real protocols, real integration patterns, real floor topology — not slideware.
