
1. Technology overview
Neysa is an AI infrastructure company founded in 2023 by Sharad Sanghi and Anindya Das and based in India, with offices in Mumbai, Bengaluru, Chennai and New Delhi. It provides GPU-based compute so that enterprises, startups and public sector bodies can train, fine-tune and run AI workloads without building their own clusters.
Velocis AI Cloud is a full-stack AI acceleration cloud. It combines GPU-as-a-Service on NVIDIA and AMD hardware (bare metal, virtual machines or Kubernetes), Inference-as-a-Service, an AI platform-as-a-Service, MLOps orchestration, real-time GPU monitoring and the Aegis LLM Shield for securing LLM endpoints. Customers named on the site include HDFC Bank, PhonePe, Juspay and the Indian Institute of Science.
2. Highlighted features & specifications
- GPU-as-a-Service: NVIDIA H100, H200, L40S and AMD MI300X capacity on bare metal, virtual machines or Kubernetes.
- Inference-as-a-Service: Managed endpoints for running models in production at low latency.
- AI Platform-as-a-Service: Managed tooling for training, fine-tuning and deploying models.
- Orchestration and MLOps: Support for Kubernetes, Slurm and vLLM based workflows.
- Real-time GPU monitoring: Visibility and cost governance across training and inference jobs.
- Aegis LLM Shield: Security layer for LLM endpoints and AI infrastructure.
3. Potential application
A Vietnamese bank can host a private LLM for customer service on dedicated GPU clusters. A university or research lab can rent reserved GPU capacity to train domain models. A software company can serve voice or copilot inference through managed endpoints.
4. Benefits
Enterprises get GPU capacity on demand, on reserved terms or as dedicated private clusters, without capital spending on hardware. The site states no egress fees and no control plane charges. Workloads can run in public cloud, private clusters, hybrid or air-gapped setups for regulated sectors. ISO/IEC 27001:2022 and SOC 2 Type II certifications are listed.
5. Market trends & opportunities
Demand for GPU compute and in-country inference is rising as enterprises move generative AI from pilots to production and data residency rules tighten. Southeast Asian banks, telecoms and manufacturers face similar needs for regional AI infrastructure outside the large global clouds.
6. Other information
Founded 2023, India. Press release dated Mumbai, 16 February 2026 announced up to 600 million USD of equity from Blackstone-affiliated funds with co-investors, plus a planned 600 million USD in debt. Earlier rounds include a 30 million USD Series A. Partnerships listed include Mavenir and Pipeshift.
One platform connecting startups, corporations, research institutions, regions and investors











Playbook & Handbook · February 6, 2026
Winning Accelerator Playbook 2026 by BambuUP
RETAIL E-COMMERCE: AI AND REAL-TIME DATA ARE TRANSFORMING HOW RETAIL BUSINESSES OPERATE








































