G&AAI: Driving Autonomous Innovation for Enterprise Transformation with Quantified Benefits
- Customer
- LTIMindtree
- Project manager on the customer side
- IT Provider
- LTIMindtree Ltd
- Year of project completion
- 2025
- Project timeline
- July, 2024 - July, 2025
- Project scope
- 10 automated workstations
- Goals
-
Reduce Manual Workload – automate routine tasks across IT, HR, and operations to let teams focus on strategic priorities.
Enhance Service Efficiency – improve speed and accuracy of service delivery.
Enable Scalable Operations – manage growth and complexity without increasing headcount.
Support AI-Driven Decision Making – provide real-time insights and autonomous agents for faster, data-backed decisions.
Deliver Conversational User Experience – intuitive natural language interfaces for seamless interaction.
Align with LTIMindtree Vision – drive digital transformation through intelligent, autonomous operations.
- Project Results
-
The GenAI and Agentic AI project has transformed the way users and teams interact with enterprise systems, delivering multiple qualitative benefits:1. Enhanced user experience through conversational, natural language interfaces—reducing the learning curve and increasing adoption.2. Faster decision-making enabled by AI-generated summaries, insights, and contextual recommendations.3. Empowered workforce, with routine tasks automated, allowing employees to focus on high-value, strategic work.4. Improved service reliability and consistency, with AI agents ensuring adherence to processes and standards.5. Reputation as an innovation leader, positioning LTIMindtree at the forefront of enterprise AI adoption within the L&T Group.
Outcomes:1. Reduced incident triaging and classification time by 30–35%.2. Cut first-response time from minutes to seconds via GenAI-assisted Virtual Agent.3. Increased adoption of self-service portals by 40%.4. Reduced reliance on SMEs for content creation by auto-generating knowledge articles with contextual tagging.
The uniqueness of the project
- Used software
-
ServiceNow Ecosystem,
NowAssist Skill set,
Agentic AI Skill set
- Difficulty of implementation
-
The project faced several key constraints:Change resistance from users unfamiliar with GenAI technologies was addressed through targeted Adoption sessions and hands-on workshops.Data quality and model alignment posed a challenge in generating accurate, context-rich responses. We resolved this by integrating curated knowledge bases, setting strict governance controls, and refining prompt engineering.Security and compliance concerns around AI usage were mitigated by aligning with ServiceNow’s trusted NowAssit framework and LTIMindtree’s enterprise risk protocols.Integration with existing processes required careful orchestration; we adopted an iterative rollout strategy with agile sprints, ensuring business continuity while embedding AI.By proactively addressing these challenges, we ensured high adoption, strong performance, and minimal disruption across business functions.
- Project Description
-
Strategic Innovation for the L&T Group
Spearheaded a pioneering initiative under the L&T Group’s innovation ecosystem, leveraging ServiceNow GenAI (NowAssist) and Agentic AI to redefine enterprise service delivery across IT and HR functions.
Cutting-Edge GenAI Integration
Implemented ServiceNow Generative AI to enable natural language interactions, automated content generation, and intelligent summarization—significantly reducing manual effort and enhancing productivity.
Autonomous Service Delivery with Agentic AI
Deployed ServiceNow Agentic AI to introduce autonomous agents that proactively resolve issues, initiate workflows, and continuously learn from interactions—ushering in a new era of intelligent enterprise operations.
Multi-Module Implementation
Rolled out the solution across key ServiceNow modules including ITSM, HRSD, and SPM, driving seamless, AI-powered service experiences across departments.
Tangible Business Impact
- Achieved up to 30% improvement in service efficiency
- Reduced turnaround times
- Elevated user satisfaction
Setting a benchmark in enterprise AI adoption.
Accelerating Digital Transformation
Positioned LTIMindtree and the L&T Group as industry pioneers in practical GenAI deployment, showcasing leadership in scalable, autonomous enterprise solutions.
This initiative exemplifies the fusion of cutting-edge technology with strategic vision, delivering measurable outcomes and transforming enterprise service paradigms. The ServiceNow GenAI and Agentic AI initiative at LTIMindtree is a pioneering implementation that integrates advanced artificial intelligence into enterprise service management to drive automation, agility, and intelligent decision-making.
Project Scope
- Integrated across key ServiceNow modules: ITSM, HRSD, and SPM
- Covers primarily internal operations environments
- Deployed in phased rollouts with early pilots in Service Desk and Asset Lifecycle workflows
Technical Architecture
1. GenAI Capabilities
- Built on ServiceNow’s Now Assist (GenAI framework) using domain-specific LLMs
- Integrated with Knowledge Management, Incident Management, and Virtual Agent to:
- Auto-generate incident summaries
- Create knowledge articles from case history
- Offer real-time search and chat-based resolutions via natural language prompts
2. Agentic AI Capabilities
- Introduced ServiceNow Autonomous Agents that:
- Monitor and analyze incidents
- Automatically recommend and execute remedial actions
- Trigger multi-step workflows with/without manual intervention
3. Security & Compliance
- Enterprise-grade security with zero data sharing with public LLMs
- Aligns with ISO 27001, SOC2, and GDPR frameworks
4. Integration & Governance
- Introduced prompt management governance and feedback loops to train agents and reduce hallucination
- Used sandbox environments for AI model testing before live deployment
This project has laid the foundation for a next-gen, autonomous enterprise platform within LTIMindtree.
- Project geography
- Global
- Additional presentations:
- Global CIO Awards - 2025 GenAI and AI Agents in LTIMindtree.pdf