04 / Telecom · Edge AI · 5G / 6G

EdgeIQ

An AI-native RAN intelligence platform for Indian telecom operators — deploying rApps on O-RAN's RAN Intelligent Controller to optimize spectrum, reduce energy consumption, manage network slicing, and enable edge AI inference — without operators building internal AI teams.

Telecom DeepTech Tier-1 telco opportunity B2B Enterprise · Licensed senior telecom advisors angle

Why now. 2026 is the breakout year for AI-native telecom (WEF + TM Forum). Indian operators have 5G SA infrastructure but lack the AI/ML talent to build the rApps that extract value — this is a managed-intelligence service problem, not a connectivity problem. The Bharat 6G Alliance calls for indigenous AI-native innovation; senior telecom advisors give direct domain access.

Proof points

2026

Breakout year for AI-native telecom (WEF + TM Forum)

15–30%

Energy savings from AI-powered RAN optimization (field data)

O-RAN

Open standard unlocks vendor-neutral AI deployment

Interface Wireframe — RAN Intelligence Console

edgeiq.in / operator / metro-circle-west / ric

EdgeIQ · Metro Circle West · Near-RT RIC Live

rAppsSlicesPredictive maint.Edge computeIntent

23.4%

Energy saved across 1,840 cells this month

Sites predicted to fail in next 24–72h

99.94%

Availability on industrial-IoT slice (SLA 99.9%)

O-RAN rApp marketplace — deployed on Near-RT RIC

rApp	Function	Cells	Impact	Status
EnergySaver	Cell sleep / carrier shutdown off-peak	1,840	−23.4% kWh	Running
SpectrumOpt	Dynamic PRB allocation	1,840	+18% throughput	Running
TrafficCast	15-min demand forecasting	1,840	MAPE 4.1%	Running
SliceGuard	Per-slice bandwidth assurance	312	SLA held	Running
AnomalyWatch	KPI anomaly detection	1,840	2 alerts today	Tuning

Intent-based operations

Operator: Maintain 99.9% availability for the industrial-IoT slice at Site 47
EdgeIQ: Translating intent → 3 RAN config changes queued
   ↳ reserve 2 PRB groups · raise slice priority to P1 · pre-warm MEC inference
EdgeIQ: Applied. Projected availability 99.96% · energy cost +1.2%

Predictive maintenance — next 72h

Site 47 · PA thermal drift~18h

Site 112 · VSWR rising~40h

Site 88 · backhaul jitter~61h

Truck rolls avoided (MTD)214

Federated learning: model improvements shared across operators — no raw network data leaves the circle (India data-sovereignty).

Edge AI compute orchestration — inference placement

Workload	Placement	Latency	Cost
Industrial vision QA (steel-plant slice)	On-device	4 ms	low
Slice demand forecast	MEC	22 ms	med
Cross-circle model training	Core	batch	high

What EdgeIQ Does

O-RAN rApp marketplace — energy savings (15–30%), spectrum optimization, traffic forecasting on the RIC
Network slice intelligence — AI-managed slicing for private-5G enterprises
Predictive maintenance engine — predict cell failures 24–72h ahead

Edge AI compute orchestration — on-device vs MEC vs core inference placement
Intent-based operations — goals → automated RAN config changes
Federated learning — shared model gains, no raw data leaves the network (data sovereignty)

Competitive Landscape

Player	India-native	O-RAN rApps	Federated learning	SME accessible
Ericsson / Nokia / Huawei	✗ Global vendors	✓ Proprietary	✗	✗ Large deals only
Global RAN OEM	✗	✓	✗	✗
Mavenir / Parallel Wireless	✗ US-based	✓	✗	◑
IT-services telecom	✓	◑ Services only	✗	✗
EdgeIQ ★	✓ India-built, India-first	✓ Open marketplace	✓ Data sovereignty	✓ SaaS model

Moats

India regulatory + data sovereignty — indigenous AI mandate
Network performance dataset from operator partnerships
O-RAN open standard = no vendor lock-in → fast adoption
Domain expertise via senior telecom-industry advisor network

Target Customers

Tier-1 telco (5G SA — largest deployment) · second telco (Open RAN)
A state-owned telco (government O-RAN mandate) · a challenger telco (cost reduction)
Private 5G: large steel & auto factories
IIT research: T2CAI defence + IISc ARTPARK