/* * Author: Yogabrata Mukhopadhyay * Organization: ManjuLab * Copyright (c) 2026 ManjuLab. All rights reserved. */

Infra-Offering

AI Data Center as a Service

Rampur, Uttar Pradesh — operational 125 kW AI infrastructure for inference, hosting, and regional compute.

Current infrastructure (operational)

Live today: 125 kW AI datacenter — Rampur, UP

  • Capacity: 125 kW (live)
  • Workloads: AI inference, model hosting, ManjuLAB platform services
  • Cooling: Hybrid cooling tuned for local industrial conditions
  • Power: Grid + UPS failover + diesel generator backup

Real-world proof-of-concept for decentralized, affordable AI compute.

Expansion roadmap (2026–2028)

Phase 1 — Completed (125 kW)

Deployed, operational, cooling and power validated.

Phase 2 — Scale to 300–500 kW

Higher rack density, GPU clusters, early enterprise & SMB customers.

Phase 3 — 1 MW target (2028)

Large-scale training & inference; backbone for Brahmexa AI marketplace.

Investment & partnerships

Operating infrastructure today — not a paper concept. Seeking partners to scale 125 kW → 1 MW with GPU-first AI capacity.

What this powers

  • Education SaaS & study-room AI workloads
  • SMB agent hosting & MCP toolchains
  • CSR form-filling agent APIs
  • Regional affordable inference for underserved markets
Discuss capacity & hosting
/* * Author: Yogabrata Mukhopadhyay * Organization: ManjuLab * Copyright (c) 2026 ManjuLab. All rights reserved. */