E2E NETWORKS LIMITED

Operates a GPU cloud computing platform that rents AI infrastructure capacity to enterprises, AI startups, and international customers running large-scale model training and inference workloads. The platform is built on a three-layer stack — TIR, Jarvis Labs and an agentic framework — and is positioned as a 'sovereign AI' factory where customers can run open-source models inside their own data boundary instead of renting frontier-model APIs. As of Q1 FY27 the live fleet is around 5,100 GPUs, including 1,024 NVIDIA Blackwell B200 chips that went on revenue during the quarter, with another 1,024 B200s expected in roughly two months and B300 and Vera Rubin on the expansion roadmap. Q1 FY27 revenue was ₹1,568M, up 334% YoY and 64% QoQ, with an EBITDA margin of 75.2% as the new fleet ramped to full utilisation. Revenue mix in Q1 FY27 was India AI roughly 20–21%, international roughly 37% and the rest domestic; an international example is a $7.7M (~₹73 crore) six-month GPU cloud order from a US-based AI service provider disclosed in April 2026. Cost structure is dominated by depreciation at 68.9% of FY26 revenue, reflecting the GPU capex load (514.1% of FY26 revenue), which is funded in part by ~₹450 crore of debt outstanding.