use case #2
The World Bank's AI-powered Search
An AI-powered Enterprise Search product for World Bank employees, delivering instant, structured answers across all internal data sources — including people, projects, documents, blogs, and media — with contextual summaries and intelligent filtering.
Here’s a preview of AI-Search* — let’s explore it in more detail.
*Please note: all data shown has been modified to comply with The World Bank’s confidentiality and security policies.
*Please note: all data shown has been modified to comply with The World Bank’s confidentiality and security policies.
Project Context and Objectives:
The World Bank's existing internal search tool lacked efficiency, returning thousands of results with minimal relevance to users' specific queries.
The goal of the AI Search project was to enhance existing tool using AI to deliver more accurate, relevant, and structured answers, improving employee productivity.
By leveraging AI, the project aimed to streamline data exploration, provide precise information quickly, and offer tailored insights for various content types across the organization's vast data resources.
The goal of the AI Search project was to enhance existing tool using AI to deliver more accurate, relevant, and structured answers, improving employee productivity.
By leveraging AI, the project aimed to streamline data exploration, provide precise information quickly, and offer tailored insights for various content types across the organization's vast data resources.
Role and Contribution:
- I played a key role in the development of the AI Search tool at the World Bank. I collaborated closely with User Researchers, Data Analysts, and both Backend and Frontend teams throughout multiple iterations of the project.
- I was actively involved in designing solutions, creating wireframes and user flows, and working on content structuring to ensure that complex data was presented in a clear and accessible way.
- I conducted a competitive analysis, examining how similar solutions, like Google Search, structure their data presentation and user interface to gain insights and inform design decisions for the World Bank's AI Search tool.
- My contributions helped refine the strategic decision-making process, offering insights based on research to refine the product at each stage. My efforts helped ensure the product met user needs and provided an intuitive experience across both desktop and mobile platforms.
*Here are some examples of working process. I'll be happy to guide you through the process in details during the interview.
Process & Methodology
- The development of the AI-powered Enterprise Search was approached iteratively, with continuous collaboration between cross-functional teams, including the Product Owner, Data Analysts, Backend and Frontend Developers. The process was driven by user-centered design (UCD) principles, ensuring that every step aligned with the users' needs and pain points.
- We began with research, including analyzing industry best practices and conducting competitor analysis, such as reviewing Google’s approach to search structures. Prototypes, wireframes, and user flows were created and refined based on user feedback and testing.
- This approach enabled us to deliver a highly functional search tool that integrated AI-driven features to provide relevant, structured results and improve the overall user experience.
Results & Achievements
- We significantly improved the search experience by integrating AI to generate precise, context-aware summaries at the top of results pages. This helped users navigate large volumes of World Bank data more efficiently, reducing the need to sift through hundreds of pages.
- Across multiple design iterations, we introduced structured, templated AI responses for different content types (e.g., people, projects, events), based on user research, backend capabilities, and cross-functional collaboration. These changes streamlined the user journey and elevated clarity across both desktop and mobile views.
- Positive feedback from stakeholders and internal users confirmed the value of this redesign. As a result, we observed growing internal adoption of the AI-powered search, especially among employees conducting cross-departmental research.
- Alexander TautererAI Search Technical Product Owner at The World BankAlice has consistently demonstrated a high level of motivation and focus in her role. She has an impressive ability to quickly grasp requirements and immediately begins the implementation process. Alice is proactive in seeking clarification on any ambiguities, ensuring that she fully understands the project needs before proceeding. Her capacity to manage multiple projects simultaneously is commendable, showcasing her excellent organizational skills and dedication to her work. Overall, Alice is a valuable asset to the team, and her contributions are greatly appreciated.
Key Learnings
Interdisciplinary Problem Solving:
Working on AI Search required deep collaboration with backend engineers, data analysts, and search experts. This enhanced my ability to bridge technical and design perspectives, particularly when dealing with unstructured data at a global scale.
Designing for Information Complexity:
I learned how to design intuitive interfaces that surface relevant results from massive datasets. I developed structures and templates for various data types, balancing consistency with flexibility to serve diverse user needs.
AI-Powered Content Structuring:
Through this project, I gained experience designing responses generated by AI. I worked on shaping structured outputs for people, projects, events, and documents, based on user intent and context, enhancing overall discoverability.
Enterprise-Level Impact Awareness:
Working within a globally recognized organization like the World Bank taught me how small UX decisions can have large-scale implications, reinforcing the importance of clarity, accessibility, and performance in enterprise environments.
Working on AI Search required deep collaboration with backend engineers, data analysts, and search experts. This enhanced my ability to bridge technical and design perspectives, particularly when dealing with unstructured data at a global scale.
Designing for Information Complexity:
I learned how to design intuitive interfaces that surface relevant results from massive datasets. I developed structures and templates for various data types, balancing consistency with flexibility to serve diverse user needs.
AI-Powered Content Structuring:
Through this project, I gained experience designing responses generated by AI. I worked on shaping structured outputs for people, projects, events, and documents, based on user intent and context, enhancing overall discoverability.
Enterprise-Level Impact Awareness:
Working within a globally recognized organization like the World Bank taught me how small UX decisions can have large-scale implications, reinforcing the importance of clarity, accessibility, and performance in enterprise environments.