August 7, 2025
5 min read
Amazon Web Services
Discover how DTDC and ShellKode enhanced logistics with DIVA 2.0, a generative AI agent built on Amazon Bedrock for smarter, real-time customer support.
The DIVA logistics agent, powered by Amazon Bedrock
DTDC is India’s leading integrated express logistics provider, operating the largest network of customer access points in the country. DTDC’s technology-driven logistics solutions cater to a wide range of customers across diverse industry verticals, making them a trusted partner in delivering excellence. DTDC Express Limited receives over 400,000 customer queries each month, ranging from tracking requests to serviceability checks and shipping rates. Their existing logistics agent, DIVA, was based on a rigid, guided workflow that forced users to follow a structured path rather than engage in natural, dynamic conversations. This lack of flexibility increased the burden on customer support teams, extended resolution times, and degraded customer experience. Seeking a more flexible and intelligent assistant capable of understanding context, managing complex queries, and improving efficiency while reducing human agent reliance, DTDC decided to enhance DIVA with generative AI using Amazon Bedrock. ShellKode, an AWS Partner specializing in modernization, security, data, generative AI, and machine learning, collaborated with DTDC to build DIVA 2.0. ShellKode leverages deep industry expertise to deliver transformative AI solutions that drive innovation and operational efficiency. In this article, we explore how DTDC and ShellKode used Amazon Bedrock to develop DIVA 2.0, a generative AI-powered logistics agent.Solution overview
To overcome the limitations of the original agent, ShellKode built an advanced agentic assistant using Amazon Bedrock Agents, Amazon Bedrock Knowledge Bases, and an API integration layer. DIVA 2.0 offers a seamless conversational interface that understands and responds naturally to customer queries. Whether tracking packages, checking shipping rates, or verifying service availability, users can interact in their own words without following rigid scripts. Enhanced AI capabilities enable DIVA 2.0 to understand context, manage complex requests, and provide accurate, personalized responses, significantly improving customer experience and reducing human intervention.Architecture and workflow
The logistics agent is built with a modular, scalable architecture integrating AWS services:
- Hosted as a static website on Amazon CloudFront and Amazon S3, integrated into the DTDC website.
- User queries are processed by AWS App Runner, which runs the web application and API services.
- App Runner invokes the Amazon Bedrock Agents API to interpret user intent.
- Based on intent, the agent triggers AWS Lambda functions to interact with DTDC backend APIs: - Tracking System API: Provides real-time consignment status. - Delivery Franchise Location API: Checks service availability. - Pricing System API: Calculates shipping rates. - Customer Care API: Creates support tickets.
- Responses are processed by Anthropic’s Claude 3.0 large language model (LLM) on Amazon Bedrock to generate context-aware replies.
- A vector-based knowledge base, built from web-scraped DTDC content, internal docs, FAQs, and operational data, is stored in Amazon OpenSearch Service to provide accurate, real-time information.
- Queries and responses are logged in Amazon RDS for PostgreSQL for scalability and relational data management.
- Monitoring and auditing are enabled via Amazon CloudWatch Logs, AWS CloudTrail, and security is enhanced with Amazon GuardDuty.
- Integrating real-time data from multiple legacy systems to provide accurate tracking, rates, and serviceability.
- Training AI to understand complex logistics terminology and multi-step queries.
- Transitioning from the rigid legacy system while maintaining service continuity and preserving historical data.
- Scaling to handle over 400,000 monthly queries with consistent performance.
- Enhanced conversations and real-time data access: Natural language understanding and multi-step reasoning enable fluid conversations. Integration with DTDC APIs provides real-time shipment tracking, service availability, and rate calculations.
- Intelligent data processing and accurate FAQ responses: LLMs process complex queries into clear, actionable insights. Knowledge bases deliver precise FAQ answers, reducing wait times.
- Reduced live agent dependency and continuous improvement: Customer support query load reduced by 51.4%. Real-time analytics provide insights for ongoing refinement.
- DIVA 2.0 achieves a 93% response accuracy supporting dynamic, natural language conversations.
- Over the last 3 months: - 71% of inquiries related to consignments (256,048), 29.5% were general inquiries (107,132). - 51.4% of consignment inquiries (131,530) did not result in support tickets. - Of inquiries leading to tickets, 40% started with customer support before AI, 60% began with AI before involving support. DIVA 2.0 has significantly reduced customer support workload, allowing the team to focus on critical issues and improving operational efficiency.
- AI Agents: The Future of Business Automation and Customer Engagement
- How AI is Revolutionizing the Cryptocurrency Trading Landscape
- The Benefits of Integrating AI in Customer Service
User interaction
Users access DIVA 2.0 through the DTDC website, submitting natural language queries about shipments, serviceability, rates, and FAQs. The system handles multi-step reasoning and dynamic conversations, providing fast, accurate, and personalized responses.
Logistics agent dashboard
The logistics agent dashboard, hosted as a static website on CloudFront and Amazon S3, is accessible only to the DTDC admin team. It uses Amazon API Gateway with Lambda backend to fetch data from Amazon RDS.
The dashboard provides real-time insights into agent performance, including accuracy, unresolved queries, query categories, session statistics, and user interactions. Visualizations such as heat maps, pie charts, and session logs enable continuous improvement and rapid issue resolution.
Solution challenges and benefits
Challenges
Benefits
Results
Summary
This case study demonstrates how Amazon Bedrock can transform a traditional chatbot into a generative AI-powered logistics agent that delivers superior customer experience through dynamic, context-aware conversations. Organizations facing similar challenges can use this blueprint to modernize AI assistants while ensuring compliance and scalability. For more information about this AWS solution, contact AWS to explore implementation, pricing, and customization options.About the authors
Rishi Sareen – Chief Information Officer (CIO), DTDC A seasoned technology leader with over 20 years driving digital transformation and innovation in logistics and supply chain sectors. Arunraja Karthick – Head – IT Services & Security (CISO), DTDC Strategic IT and cybersecurity leader with 15+ years transforming legacy environments into secure, cloud-native ecosystems. Bakrudeen K – AI/ML Practice Head, ShellKode AWS Ambassador leading AI/ML innovation with expertise in generative AI and agentic assistants. Suresh Kanniappan – Solutions Architect, AWS Specializes in cloud security and industry solutions for automotive, manufacturing, and logistics. Sid Chandilya – Sr. Customer Relations Manager, AWS Expert in tech-led business transformation and AI-powered customer experience.Source: Originally published at Amazon Web Services Blog on 07 Aug 2025.