Inventory & lifecycle
Provision, group, tag and retire devices at scale. Track firmware, health and ownership across every hardware class from one console.
GPU-first edge orchestration
One platform for every device at the edge.
MeshLogiq unifies microcontrollers, edge compute, IP cameras and robots into a single secure control plane — with digital twins, real-time telemetry, and GPU-accelerated video analytics built in.
4+
Device classes
GPU-first
NVDEC · CUDA · TensorRT
Multi-tenant
Isolated by org
1,284 devices online
Live analytics · 28ms
Built to manage heterogeneous fleets
ESP32 & MCUs
Jetson & Raspberry Pi
RTSP / ONVIF cameras
ROS2 robots & arms
The platform
Everything you need to run a connected fleet
From onboarding a single sensor to orchestrating thousands of robots, MeshLogiq gives your team one operational surface for the entire device lifecycle.
Digital twins
Model desired vs. reported state with full history. Reconcile drift automatically and replay any device's timeline on demand.
Telemetry & events
Ingest high-frequency metrics and events over MQTT into a schema built for growth — indexed, queryable and ready for analytics.
Cloud → device commands
Send secure, auditable commands and config to any endpoint — individually or across whole fleets — with delivery guarantees.
Video & event clips
Stream live RTSP/ONVIF feeds and capture event-driven clips automatically, triggered by on-server analytics in real time.
Multi-tenant by design
Organizations, users, roles and permissions with strict per-tenant data isolation — so teams and customers never see each other.
GPU-accelerated core
Every server ships with NVIDIA GPUs — and we use them.
MeshLogiq is engineered GPU-first. Video decode, encode and inference run on-device in the datacenter, so real-time analytics scale without melting your CPUs.
- NVDEC / NVENC hardware decode & encode for dozens of concurrent camera streams per node.
- CUDA & TensorRT inference pipelines for low-latency detection, tracking and event triggers.
-
Verified container GPU access via the NVIDIA Container Toolkit —
--gpus all, production-tested.
NVIDIA · GPU verified
32×
Concurrent streams / node
<30ms
Inference latency
NVDEC
Hardware decode path
TensorRT
Optimized models
Any device, any protocol
Heterogeneous fleets, one control plane
Stop stitching together vendor tools. MeshLogiq speaks the languages of the edge and normalizes them into a consistent model.
- ESP32 & microcontrollers — lightweight MQTT agents for constrained IoT hardware.
- Jetson & Raspberry Pi — edge compute nodes with local processing and sync.
- RTSP / ONVIF cameras — discovery, streaming and server-side analytics.
- ROS2 robots & arms — command, monitor and coordinate autonomous systems.
Operations console
See your whole fleet at a glance
Live device status, telemetry, video and analytics — in one responsive, real-time dashboard.
Secure by default
Enterprise-grade security at every layer
Least-privilege access, strong identity and encrypted transport are baked in — not bolted on.
Keycloak identity
Centralized authentication and authorization with SSO, roles and fine-grained permissions per organization and device.
TLS via Traefik
Network-level encryption everywhere with automated certificates and secure routing at the edge of your infrastructure.
Least privilege
Every device and user gets exactly the access it needs — scoped credentials, isolated tenants, full audit trails.
Bring order to the edge
Connect your first device in minutes and scale to thousands. See how MeshLogiq unifies IoT, cameras and robotics on one GPU-first platform.