The TAMM platform, Abu Dhabi Government’s exclusive one-stop shop, offers more than 940 services to citizens, residents, visitors, and businesses in the emirate. Recognised globally for its cutting-edge technologies and innovation, TAMM enables users to access essential services such as applying for driving licenses, paying fines, and retrieving personal documents.
In October 2024, during GITEX Global, the Department of Government Enablement – Abu Dhabi (DGE) launched the third-generation evolution of the platform, TAMM 3.0, to its more than 2.8 million active users. This launch highlighted the growing role of artificial intelligence (AI) within the TAMM ecosystem, demonstrating that AI now embedded every aspect of TAMM – from service delivery to customer care, to day-to-day operations and decision-making. At the core of this transformation is the AI-powered TAMM Assistant, which has become a game-changer in service delivery.
TAMM’s AI Assistant goes beyond answering questions – it proactively anticipates customer needs, performs personalised tasks, automates service execution, and generates actionable insights to create a seamless government experience. Built on LangChain and LangGraph, this advanced system is a major step forward in enhancing citizen engagement and optimising government services.
Agentic Framework – Five Key Workflows for TAMM’s AI Assistant
The TAMM AI Assistant is powered by LangChain, a system chosen for its proven flexibility and robust capabilities in seamless handling of complex workflows. Guided by the DGE and TAMM leadership, TAMM’s AI engineering team prioritised creating an efficient and user-centric system, enabling seamless access to TAMM’s broad range of services. From personal document inquiries to support requests, every interaction reflects DGE’s commitment to innovation and operational excellence.
LangChain provided the necessary infrastructure to manage complex queries, retrieve relevant data, and facilitate seamless interactions across multiple systems. At the heart of TAMM’s AI Assistant are five critical workflows that ensure government services are accessible, personalised, and streamlined:
1) Service inquiry and information retrieval: This flow enables the assistant to efficiently address a wide variety of questions related to TAMM’s 940+ services. By utilising an advanced RAG pipeline, powered by LangChain, the assistant can connect to TAMM’s service databases to provide accurate and concise responses. It can answer questions such as:
- “How do I renew my passport?”
- “What documents are required for a driving license application?
- What’s the weather today in Abu Dhabi?
2) User-specific data interaction: Designed to provide personalised service, this flow tailors responses based on the user’s individual data and history. It allows the assistant to address user-specific queries such as:
- “Show me my valid documents.”
- “What payments do I owe?”
- “What documents are about to expire?
3) Service execution: This flow enables users to seamlessly execute commonly requested services directly through the AI Assistant. Rather than recreating services within the assistant, TAMM’s ecosystem is channel-agnostic, allowing users to access and complete services across any platform. Whether using the mobile app, web portal, or AI Assistant, users enjoy a consistent and unified experience.
4) General knowledge and discussion: This flow harnesses the broad capabilities of
Large Language Models (LLMs), enabling users to expand their enquiries beyond specific topics. For instance, users can ask questions like, “What is the best place to go camping in Abu Dhabi?” and receive insightful, context-aware responses.
5) Support flow: This flow is designed to direct users to the appropriate support channel based on their intent. A standout feature in TAMM 3.0 and the TAMM AI Assistant is "Take Photo and Report," which enhances the ability of the Abu Dhabi Government to assist users with their everyday needs. This feature allows users to capture a photo or video of any incident. The AI then extracts the relevant details, automatically submitting the report with time, date, and location stamps. The incident is then routed to the appropriate Abu Dhabi Government entity for resolution, ensuring a streamlined and efficient response process.
Additionally, TAMM AI Assistant operates within strict guardrails to ensure that all responses adhere to the standards set by the Abu Dhabi Government.
Leveraging LangChain’s robust orchestration capabilities, the AI Assistant seamlessly integrates with TAMM’s unified backend, enabling real-time functionality for users without adding unnecessary complexity.
Together, these advanced workflows strengthen TAMM’s position as a global leader, offering a unified, intuitive super app experience across multiple channels. Whether for citizens, residents, visitors, or businesses, TAMM continues to lead the way in delivering a comprehensive, omnichannel service experience.
Why LangGraph? Elevating Flexibility and Standardisation
TAMM’s AI engineering team designed TAMM’s AI Assistant to scale with the growing needs of its users. Leveraging LangChain and LangGraph, the system transitioned to a multi-agent architecture that ensures agility, efficiency, and precision in managing user data including payments and document histories. This forward-thinking approach underscores TAMM’s role, under DGE, as a leader in digital government services.
What’s Next?
TAMM’s AI Assistant represents the future of citizen engagement, with advanced coordination among AI agents ensuring seamless and context-aware service delivery. This innovation reflects DGE’s unwavering commitment to setting global benchmarks in digital transformation.