use case
The World Bank's AI-driven Chat
Problem Statement
- Sensitive Data RiskEmployees started using external AI tools in their daily work, creating potential risks around sharing confidential World Bank data outside secure environments.
- Growing Demand for AIEmployees began exploring how AI could support their workflows, showing a clear interest in integrating AI into everyday tasks.
- Demand for Unified AI Apps LayerAI solutions started to emerge across different teams, but remained isolated, highlighting the need for a unified space to access and reuse them.
- Demand for Integrated Document AccessEmployees needed a way to work with internal documents in a secure environment, with the ability to explore and interact with enterprise knowledge directly through chat.
Product Vision
- New Interaction ModelWe are introducing a conversational layer that redefines how employees interact with institutional knowledge — moving from fragmented search across systems to a single, natural language entry point.
- Unified Knowledge ExperienceInstitutional knowledge becomes accessible through a unified AI layer that can understand context, retrieve relevant information, and synthesize insights across multiple internal sources.
- Trust-First AI EnvironmentThe system is designed with embedded governance, ensuring that sensitive data, access rights, and responsible AI principles are enforced at every step of interaction.
- Organizational Capability ShiftThe product enables a shift from manual information discovery to AI-assisted decision-making, improving speed, consistency, and quality of work across the organization.
MVP (Minimum Viable Product )
The MVP introduced a simple conversational interface to explore how employees interact with AI in a secure environment. Users could ask questions, receive AI-generated responses, based on a limited set of internal sources, including curated collections of documents and reports, projects.
The product was built using OpenAI model deployed via Microsoft AzureDevOps, ensuring that sensitive institutional data was not exposed or reused for external model training.
MVP UX Research
MVP was rolled out to a small group of internal teams and collected feedback through internal platforms like MS Viva Engage, as well as direct outreach by UX researchers. In parallel, moderated usability sessions were conducted via MS Teams video calls, where users were given simple tasks and observed during their first interaction with the product. This helped us capture both user feedback and real behaviour, forming the foundation for our key learnings.
- Anastasia Dovgaya I Product Manager (Ex - UX Researcher) at The World Bank"Working closely with the UX research team, Alice attended all interview sessions with participants, took into consideration recommendations from UX researchers, and quickly addressed pain points identified during the study. It was the joy, delight, and such a great experience to work with her."
MVP Learnings → Product Decisions
- 1. PromptingUsers experienced difficulties structuring prompts, which led to inconsistent output quality.
→ This led to the introduction of a prompt library designed to help users quickly select and reuse effective prompts instead of writing them from scratch. - 2. Onboarding / WalkthroughFirst-time users were often unclear about product capabilities and interaction patterns.
→ A guided onboarding experience was designed to help users quickly understand key functionalities and usage flow. - 3. File & Image InputThere was a strong expectation for the ability to upload and analyze files and images directly within the chat.
→ This led to prioritization of multimodal input capabilities to support document- and image-based workflows. - 4. Conversation HistoryUsers faced challenges in locating and reusing previous conversations and outputs.
→ This led to a redesign of conversation history into a more structured and searchable knowledge memory layer. - 6. Knowledge FilteringUsers needed the ability to narrow responses to specific domains and internal datasets.
→ This resulted in the introduction of structured knowledge filtering based on organizational context such as projects and datasets. - 7. Feedback LoopAlthough users were willing to provide feedback, the mechanism for doing so was not intuitive or easily accessible.
→ This led to embedding contextual feedback mechanisms directly within the chat experience. - 8. Multiple Model ExpectationsUsers expressed interest in using different AI models depending on task type and desired output style.
→ This led to the exploration of a flexible model layer allowing users to switch between different AI providers within the same chat experience. - 9. Translation NeedsUsers frequently worked with documents and workflows in multiple languages, which required constant switching to external translation tools.
→ This led to the consideration of in-chat translation capabilities to reduce reliance on external services.
Main MVP Insight:
AI was expected to support end-to-end work processes rather than serve purely as a conversational assistant.
→ This insight informed a strategic shift toward building a unified AI productivity layer across enterprise knowledge and workflows.
AI was expected to support end-to-end work processes rather than serve purely as a conversational assistant.
→ This insight informed a strategic shift toward building a unified AI productivity layer across enterprise knowledge and workflows.
Product 1st version (build on MVP)
- Need more functionality (as Upload Image)Bla bla We are a leading firm in providing quality and value to our customers. Each member of our team has at least 5 years of legal experience. We like what we do.
- No need in follow upsBla bla Each member of our team has at least 5 years of legal experience. They use their knowledge to make our clients' lives better.
- Need AppsBla bla Our managers are always ready to answer your questions. You can call us at the weekends and at night. You can also visit our office for a personal consultation.
- Need to connect it to Enterprise SearchBla bla Our company works according to the principle of individual approach to every client. This method allows us to achieve success in problems of all levels.
- Need access to Enterprise DocumentsBla bla Our prices are fixed for some standard services and we offer discounts for regular clients. Also, we ask our new clients about their birthday and prepare cool presents.
- Need Instructions and Whats new and Product Guide popupsBla bla Our prices are fixed for some standard services and we offer discounts for regular clients. Also, we ask our new clients about their birthday and prepare cool presents.
Product Evolution
System & Architecture
Product Thinking
UX & Design System
Delivery Process
Impact & Metrics
My Role
- The project involved the development of a secure, closed chat system for The World Bank users, specifically designed to integrate with The World Bank’s Data and Documents Repository.
- The goal was to create an internal chat solution that allows users to access and query sensitive data without using open-source platforms such as ChatGPT, Perplexity, Gemini, and others, to avoid potential data leaks.
- The chat system aimed to provide a user-friendly interface for retrieving data in a conversational manner, ensuring both security and ease of use while maintaining the integrity of sensitive information.
Role and Contribution:
- As a key contributor to this project, I closely collaborated with the Product Owner, technical specialists (backend and frontend developers), user researchers, business analysts, and data researchers.
- I was responsible for conducting user research, which laid the foundation for creating detailed user personas. These personas were then used to identify specific user needs, which guided the development of empathy maps and user flows. Following this, I created wireframes to visualize the user experience, which were further refined into high-fidelity designs.
- Finally, these high-fidelity designs were transformed into animated prototypes for a comprehensive user experience. I also participated in presentations, effectively communicating our progress and designs to stakeholders.
Process & Methodology
- As a central part of the team, I collaborate with the Product Owner, Scrum Master, back-end developers, data researchers, and other specialists to define user needs and align on project goals. Based on collected insights, I develop User Personas, User flows, User Journey Maps, wireframes, high-fidelity designs and animated prototypes.
- I also create and maintain the project roadmap, setting deadlines, identifying key milestones and challenges, and helping the Product Owner write user stories to ensure alignment with user and business objectives.
Results & Achievements
- We began with the core functionality of matching ChatGPT, aimed at introducing users to the fundamental features of AI-driven chat applications. As user engagement grew, we identified new opportunities to expand the functionality, and over four iterations, we introduced advanced features based on market research and user feedback. These updates included enhanced usability and new tools, like document and image analysis, which significantly improved the overall user experience.
- Feedback from key stakeholders, including the Product Owner, highlighted the success of these updates. Positive user reviews further reinforced the value of these improvements, driving increased adoption. As a result, we’re seeing a growing number of users engaging with the platform regularly.
- We continue to gather insights from user interactions and plan future updates to address additional needs, ensuring the product evolves alongside its users’ expectations.
Key Learnings
- Interdisciplinary Collaboration: I learned to collaborate closely with a diverse range of specialists — developers, data researchers, product owners, and more — within a globally recognized organization like the World Bank. This enhanced my communication and coordination skills across cross-functional teams, dealing with a variety of expertise and perspectives.
- Flexibility in Design Process: I gained a deep understanding of the importance of adaptability in the design process. I had to quickly adjust and rethink initial decisions based on evolving user needs, new functionalities, and feedback within the global context of the project.
- Adapting to New Requirements: I developed the ability to swiftly adapt to changes in project requirements, including the introduction of advanced features and functionality, ensuring that the product always met the dynamic needs of World Bank’s users.
- Data-Driven Decision Making: I learned to leverage real user data to inform design decisions, ensuring that the product was aligned with user expectations, and optimized for the needs of a global audience.
- AI & Machine Learning Integration: I deepened my understanding of how AI and machine learning technologies can be integrated into user experience design to create smarter, more adaptive features that can scale to meet the demands of a global enterprise.
- Project Management & Roadmap Development: I refined my skills in creating and maintaining project roadmaps, managing progress, and identifying potential risks, while aligning timelines and expectations with stakeholders from a worldwide organization.
