India’s Global Systems Integrators Build Next Wave of Enterprise Agents With NVIDIA AI, Transforming Back Office and Customer Support
Source: NVIDIA AI Blog
Agentic AI Transforming India’s Tech Landscape
India’s technology sector is leveraging NVIDIA AI Enterprise and NVIDIA Nemotron models to boost productivity and efficiency across a wide range of industries—from call centers and telecommunications to healthcare.
Key Players
- Infosys
- Persistent
- Tech Mahindra
- Wipro
These firms are spearheading business transformation by integrating agentic AI platforms built on NVIDIA AI Enterprise, enhancing back‑office operations and customer‑service experiences.
Highlights from the India AI Impact Summit
- The 2024 India AI Impact Summit showcased next‑generation business services powered by agentic and generative AI.
- Demonstrations highlighted real‑world use cases and the tangible benefits of AI‑driven automation.
Market Outlook
- According to the India Brand Equity Foundation (IBEF), the Indian tech industry is projected to reach $500 billion in revenue by 2030, up from roughly $250 billion in 2023.
- This growth is fueled by a surge in AI adoption, exemplified by the procurement of 38,000 GPUs in September alone.
For more information on the summit:
- Visit the official site: India AI Impact Summit
Wipro WEGA Platform – Boosting Call‑Center Efficiency with NVIDIA AI Enterprise
Overview
Wipro’s AI‑agent‑assisted solution, built on the WEGA platform and NVIDIA AI Enterprise, is transforming how a major U.S. health‑insurance provider serves its members. By combining real‑time AI assistance with a centralized data hub, the system enables agents to handle more complex requests, cut resolution times, and deliver personalized, 24/7 support.
Business Challenge
| Issue | Why It Matters |
|---|---|
| Seasonal hiring & long training cycles | Limits the ability to scale quickly during peak enrollment periods. |
| Rising call volumes & fragmented data | Increases wait times and puts pressure on human agents. |
| Heavy administrative workload | Drains resources that could be used for higher‑value interactions. |
| Regulatory compliance | Requires strict governance and data‑privacy safeguards. |
Solution Architecture
- Core Platform: Wipro WEGA (Enterprise‑grade contact‑center platform).
- AI Stack: NVIDIA AI Enterprise suite, featuring:
- NVIDIA NIM microservices – production‑grade, horizontally scalable inference services.
- NVIDIA NeMo Guardrails – built‑in safety and compliance controls for regulated domains.
- Key Capabilities:
- AI‑agent assistance – real‑time prompts, knowledge retrieval, and conversational self‑service.
- Centralized data hub – aggregates member data to surface personalized insights.
- Automated digitization – removes manual steps from downstream processes.
Note: All components run with sub‑200 ms latency, supporting up to 900 concurrent calls and 164 requests per second.
Results
- 42 % of inbound calls now handled entirely by AI agents.
- Near‑instant responsiveness across 900 concurrent calls.
- Throughput of 164 requests / second.
- NVIDIA NeMo Guardrails ensure compliance and safety.
Wipro’s AI‑agent‑assisted solution demonstrates how combining a robust contact‑center platform with NVIDIA’s AI Enterprise can deliver faster, more accurate, and compliant customer experiences—setting a new standard for the industry.
Tech Mahindra Deploys Large Telco Model (LTM) to Power Autonomous Network Operations Using NVIDIA NIM
Tech Mahindra is accelerating AI‑assisted network operations with a new platform built together with NVIDIA. At its core is a Large Telco Model (LTM) that generates prioritized, data‑driven recommendations for field technicians, ranking each fix by its historical success rate across the network. The result: faster, more accurate resolutions—often in a single visit—and a clear path toward Level‑4‑plus operational maturity.
Key Benefits
- Higher first‑time‑fix rates → fewer repeat visits.
- Improved service‑layer issue resolution → quicker restoration of services.
- Enhanced customer experience → reduced churn and higher satisfaction scores.
- Back‑office efficiency → higher‑quality tickets and fewer escalations.
How the Platform Works
| Component | Role | NVIDIA Technology |
|---|---|---|
| Semantic Search | Indexes and retrieves relevant telemetry, logs, and documentation. | NVIDIA Nemotron embedding models |
| Reranking | Refines search results to surface the most actionable insights. | Nemotron reranking model |
| Inference Engine | Serves the models at scale with low latency. | NVIDIA NIM micro‑services |
| Workflow Orchestration | Coordinates agent actions across network domains. | NVIDIA NeMo Agent Toolkit |
Deployment Highlights
- Micro‑service architecture using NVIDIA NIM enables rapid, reliable AI inference.
- Agentic operations are orchestrated by the NeMo Agent Toolkit, allowing autonomous decision‑making across multiple network domains.
- Scalable: Designed to support global telecom operators handling > $1.5 trillion in annual revenue, where even modest uptime gains have massive economic impact.
Industry Impact
By embracing autonomous network operations, Tech Mahindra demonstrates how AI can transform the telecommunications sector—delivering measurable uptime improvements, operational cost reductions, and a competitive edge in a market worth over $1.5 trillion annually.
Learn more about the underlying models: NVIDIA Nemotron Foundation Models
Infosys Builds an Enterprise‑Grade Coding Small Language Model with NVIDIA AI Enterprise
Infosys has introduced a 2.5‑billion‑parameter small language model (SLM) for software development. The model is built on the NVIDIA NeMo framework—part of NVIDIA AI Enterprise—and is integrated into Infosys Topaz Fabric. It delivers frontier‑grade performance while remaining lightweight enough for deployment on:
- On‑premises enterprise data centers
- Public‑cloud environments
- Standard desktop machines
Key Capabilities
| Capability | Description |
|---|---|
| Agent development | Enables creation of AI‑driven coding assistants and multi‑agent pipelines. |
| Code generation | Produces syntactically correct code snippets across multiple programming languages. |
| Refactoring | Suggests structural improvements and optimizations for existing code bases. |
| End‑to‑end software‑engineering workflows | Supports the full lifecycle—from design and implementation to testing and debugging. |
Training Data & Performance
- Data mix: Curated high‑quality code repositories, synthetic code, mathematical‑reasoning datasets, and natural‑language inputs.
- Benchmark results: Matches frontier‑model performance on MBPP, MBPP+, and BFCL benchmarks.
Safety, Trust, and Secure‑Coding
- Safety‑aligned training: Incorporates responsible‑AI practices to reduce harmful or biased outputs while preserving fluency.
- Secure‑coding validation: Tested against industry benchmarks such as Stanford AIR‑Bench and Meta’s CyberSecEval, giving enterprises confidence in its security‑focused code suggestions.
Typical Use Cases
- Automated code generation for routine tasks
- Intelligent debugging and error‑resolution assistance
- Refactoring legacy codebases with AI‑driven recommendations
- Building multi‑agent development pipelines for complex software projects
Infosys’ coding SLM demonstrates how enterprise‑grade AI can accelerate software delivery without sacrificing security or trust.
Persistent Accelerates AI‑Driven Molecular Discovery with NVIDIA BioNeMo and NeMo Agent Toolkit
Persistent Systems is partnering with NVIDIA to usher early‑stage drug discovery into a new era of speed and scientific fidelity. By combining Persistent’s deep life‑sciences engineering expertise with NVIDIA’s full‑stack accelerated‑computing platform, researchers gain a powerful path from AI experimentation to production‑grade discovery workflows.
What’s Being Delivered
- GenMoIVS (Generative Molecules and Virtual Screening) – a solution built on the NVIDIA BioNeMo platform and the NeMo Agent Toolkit.
- Domain‑specific large models that simulate molecular behavior with high accuracy, generating and evaluating candidate compounds before they ever reach a wet lab.
- Agentic workflows that continuously reason across virtual screening, prioritization, and experimental planning, helping teams de‑risk early discovery and shorten development cycles.
Technical Foundation
- Runs on NVIDIA’s accelerated‑computing stack, including NVIDIA AI Enterprise software and NIM micro‑services.
- Enables high‑throughput simulation and real‑time scientific decision‑making in regulated environments.
- Scalable infrastructure paired with production‑ready agentic AI.
Why It Matters
- Faster exploration of compound space – AI‑driven generation and screening dramatically cut the time needed to identify promising candidates.
- Cost‑effective discovery – Reduces the number of wet‑lab experiments required, lowering overall R&D spend.
- Higher downstream success rates – More accurate early‑stage predictions improve the likelihood of later‑stage success.
By merging scalable hardware, sophisticated AI models, and production‑grade tooling, Persistent and NVIDIA are giving life‑sciences organizations a faster, more reliable way to discover new therapeutics.