IDB Knowledge Management Platform Development TORs 2024 PDF

TERMS OF REFERENCE PROPOSAL FOR THE DEVELOPMENT OF THE KNOWLEDGE MANAGEMENT PLATFORM OF THE INTER-AMERICAN DEVELOPMENT BANK (IDB) November 2024 1. Background y Justificación......

TERMS OF REFERENCE PROPOSAL FOR THE DEVELOPMENT OF THE KNOWLEDGE MANAGEMENT PLATFORM OF THE INTER-AMERICAN DEVELOPMENT BANK (IDB) November 2024 1. Background y Justificación....................................................................................... 1 3. Users.................................................................................................................... 5 4. Scope of Work....................................................................................................... 7 5. Key Deliverables..................................................................................................... 8 6. Key Documentation............................................................................................... 11 7. Project Schedule and Milestone.............................................................................. 11 8. Reporting Requirements........................................................................................ 11 9. Acceptance Criteria............................................................................................... 13 10. Eligibility.............................................................................................................. 13 12. Supervision.......................................................................................................... 14 Background & Justification The Inter-American Development Bank (IDB) is the main source of multilateral financing for economic, social, and institutional development in Latin America and the Caribbean. It provides loans, grants, guarantees, policy advice, and technical assistance to the public and private sectors of its borrowing member countries. The Knowledge and Learning Division (KLD) within the Vice Presidency of Sectors and Knowledge (VPS) of the IDB, is responsible for: (i) helping the IDB Group to identify, build, apply, and disseminate high-quality knowledge, as well as to develop and use learning products that enable the building of technical and general competencies; (ii) promoting an internal culture of innovation by providing experience, networks, and methodologies, and expediting the adoption of innovative solutions; and (iii) expanding the audience that uses the Bank’s products and positioning the brand through internal and external communication. Within KIC, the Knowledge and Learning Division (KIC/KLD) provides clients with guidance and tools to capture, generate, manage, and use knowledge, and supports the planning and development of learning products. In the context of preparing the new Institutional Strategy of the IDB, as well as the diagnostics and proposals that have been developed simultaneously related to knowledge and effectiveness, KLD has formulated an ambitious work program. This program aims to support our mandate and the IDB Group’s vision of consolidating the Bank as the region’s "Knowledge Bank." To achieve this vision and these objectives, the KLD team will focus its knowledge management and capacity- building efforts on the following priorities: Enhancing the quality of our knowledge products. Fully harnessing the virtuous cycle between operations and knowledge. In this context, the IDB seeks to develop a robust knowledge management platform, The IDB Group Knowledge Platform is an internal digital solution powered by generative artificial intelligence (Gen AI) that facilitates an easy access to crucial information, promoting the generation, systematization, and use of knowledge, unlike existing solutions that present disconnected and fragmented knowledge. This project also includes the development of a lessons-learned writing assistant, LessonsGPT AI writer assistant, aimed at facilitating the capture and systematization of knowledge within the organization. This initiative aligns with the IDB’s vision of establishing itself as the "Knowledge Bank" in Latin America and the Caribbean, providing a comprehensive solution to organize, systematize, and facilitate the use of knowledge generated within the institution. Product Context The Knowledge Management Platform aims to bridge gaps between analytical and operational knowledge through efficient search, systematization, and interaction, making knowledge easily accessible within the IDB Group. In the initial phase, the platform will be accessible exclusively to an internal audience. The project consists of two MVPs, each with specific capabilities and objectives: Knowledge Platform - Capability "Search and Interact": o MVP Objective: Implement project-guided search, allowing the platform to recognize projects associated with the team leader and offer structured knowledge in specific categories. This will facilitate access to information related to knowledge products and operational documents, such as documents of similar projects, institutional documents of the IDB, relevant literature from IDB publications and external sources, and operational documents with lessons learned. o Current Phase: ▪ Non-functional Prototype: A non-functional prototype has been tested with 20 users to gather feedback (Annex 1).  ▪ Aurora - Operational Knowledge Assistant: Aurora, developed under the SaaS model, uses a RAG (Retrieval-Augmented Generation) approach to enhance the use of artificial intelligence and large language models (LLMs). This approach enables the platform to retrieve relevant information from a set of documents before generating responses, ensuring that the AI’s answers are based on specific and accurate data. Through RAG, Aurora’s language models can access directly to operational and up-to- date information, thereby improving the accuracy and relevance of the responses provided to IDB experts. ▪ Success Criteria: The Knowledge Platform product team has already defined the necessary metrics to evaluate the platform’s effectiveness and performance, ensuring continuous monitoring of its impact and value for users. ▪ User Archetypes: The team is deepening the research phase to understand in detail the user archetypes for the Knowledge Platform. This enrichment process seeks to map the needs and behavior patterns of different users within the Bank, from project leaders to operations analysts. ▪ Reference Architecture: The reference architecture establishes key components, their interactions, and the infrastructure needed to ensure scalable, secure implementation aligned with best practices. Through this reference, the development team will be able to understand data flows and integration points at a high level (Annex 2). Lessons GPT - Capability "Capture and Systematize Lessons Learned": o MVP Objective: Implement the Lessons GPT tool to capture, structure, classify and store evidence in the form of lessons learned. In the first release, this tool will be an independent component but with the ability to connect via APIs to the Bank's systems. o Current Phase: The product is currently in the proof-of-concept stage, with the following work already completed as a foundation for future development: ▪ Model Training: A database of approximately 400 'Strong' and 400 'Weak' lessons learned was used to train the model, ensuring the generation of actionable, robust, and structurally sound lessons through iterative refinement of information-gathering questions. ▪ Generative AI Capability: Generative AI was utilized to process user inputs and provide coherent, contextually relevant lessons learned, streamlining the knowledge generation process while ensuring data privacy  through Azure OpenAI, fostering user confidence in sharing detailed information. ▪ User Testing: Extensive testing was conducted with IDB Project Team Leaders, Operations Analysts, and Project Executing Units (PEUs), resulting in strong user acceptance and valuable feedback on question framing, lesson structure, and overall tool usability. ▪ Use Cases: The Bank’s internal product design team is evaluating various use cases for Lessons GPT. This analysis will help identify how the generative AI assistant can effectively integrate into current workflows, adding value to the structuring and generation of lessons learned. This input is crucial for advancing the construction of the prototype. ▪ Non-functional Prototype: A functional prototype of the tool has not yet been developed. How We Work: Our team operates aligned with the Inter-American Development Bank (IDB) Group's New Operating Model (NOM), focusing on building a digital product ecosystem for the Knowledge Platform with a product-oriented perspective. We work as a multifunctional team, integrating Business, Design, and Technology, and strategically consider how each of our contributions impacts our users' experience. 1. Objectives The objective of this contract is to develop and deploy two parallel products, Knowledge Platform and LessonsGPT AI writer assistant, which will fulfil specific and complementary functions in knowledge management within the IDB. Both products will be developed as MVPs, with a modular and scalable architecture that allows their evolution in future development cycles, leveraging large language models (LLMs) to facilitate the search, systematization, and interaction with institutional knowledge. 1. Knowledge Platform: A knowledge search and interaction platform within the IDB, using the *Aurora* search engine. This MVP will enable users to access knowledge to meet operational needs through data-driven interactions. Specific objectives of Knowledge Platform: - Develop and implement a functional MVP, meeting high standards of quality and user experience, allowing for initial testing and validation.  - Design a modular and scalable architecture that allows the continuous evolution of the platform, incorporating new use cases and functionalities, and adapting to changing requirements with minimal rework. - Create medium and high-fidelity prototypes for new components and functionalities, subjecting them to user testing to gather feedback that will refine the design and enhance the platform’s value proposition. - Implement a user-centred design that prioritizes users' needs and preferences, ensuring an intuitive, accessible, and personalised experience, with customization options and adjustments based on continuous feedback. 2. LessonsGPT AI writer assistant: A generative AI assistant for the structured and contextualized creation of lessons learned, enabling effective capture of organizational knowledge. Specific objectives of LessonsGPT: - Develop a user-centered interface powered by generative AI to assist team leaders and other operational stakeholders in drafting structured and comprehensive lessons learned throughout the project lifecycle, with targeted questions to optimize the process. - Offer text and voice features to respond to targeted questions. - Organize, classify, and store lessons learned for easy dissemination and reuse in future operations. - Enable the functionality to copy the results of lessons learned. - Enable the admin of the tool to configure what questions/steps will be asked the user to complete and what information will be retrieved from other systems to enhance the results 3. Cross-cutting objectives for both products: - Ensure that the proposed architecture is scalable and secure, capable of handling multiple concurrent users while protecting user interactions and data privacy. - Ensure that the MVP of LessonsGPT is an independent component with the ability to connect via APIs to the IDB’s systems, enabling smooth integration with the existing technological ecosystem. Users The knowledge platform is an internal digital solution of the IDB Group. In the product's first phase, the product design team prioritized sector specialists who design sovereign-guaranteed operations, known as Project Team Leaders (PTLs). There are additional use cases to explore with Sector Specialists who are project team members but not PTLs, as well as Operations Analysts, REAs, SEAs, Country Economists, Knowledge Coordinators, Effectiveness Focal Points, and other users from strategic areas. Project Team Leader (PTL): The person responsible for leading and managing a project team. Here are the key responsibilities: (i) Project Planning: Developing detailed project plans, including timelines, budgets, and resource allocation; (ii) Team Coordination: Leading and coordinating the project team, ensuring that all members are aligned with the project goals and timelines; (iii) Stakeholder Engagement: Communicating with stakeholders, including government officials, partners, and beneficiaries, to ensure their needs and expectations are met; (iv) Monitoring and Evaluation: Overseeing the monitoring and evaluation of project progress, ensuring that objectives are being met and making adjustments as necessary; (v) Reporting: Preparing and submitting regular progress reports to the IADB and other stakeholders. Operation Analyst: Supports project planning, implementation, and closure. Alternate PTL and team members: sectoral specialists from VPS, SPD, and FMP, help develop the loan proposal and support the Team Leader (TL) during the implementation phase in their respective areas. Regional Economic Advisor (REA): An expert who provides economic analysis and advice on regional economic issues to support the IDB’s operations and strategies. Their key responsibilities include: (i) Analyzing regional economic trends: monitor and evaluate economic developments across multiple countries within a region to identify trends and potential impacts on the IDB’s operations, (ii) Advising on regional policies: provide guidance on evidence-based economic policies and strategies that can enhance regional integration and development, (iii) Supporting project teams: collaborate with project teams to ensure that regional economic considerations are integrated into project design and implementation, and (iv) Facilitating knowledge sharing: promote the exchange of economic knowledge and best practices across countries in the region to address common challenges and opportunities. Sector Economic Advisor (SEA): An expert who offers economic insights and guidance on specific sectors, such as infrastructure, education, or health, to inform the IDB’s projects and policies. Their responsibilities include: (i) Sector-specific analysis: Conducting in-depth economic research and analysis within their designated sector to inform IDB projects and policies, as well as coordinating the knowledge agendas of the sector, (ii) Policy advice: Offering expert guidance on sectoral policies and strategies to enhance the effectiveness and impact of IDB interventions, (iii) Supporting project design: Collaborating with project teams to ensure that sectoral economic considerations are integrated into project planning and implementation, and (iv) Knowledge dissemination: Sharing sector-specific knowledge and best practices to promote learning and innovation within the IDB and among its stakeholders. In the case of LessonsGPT AI writing assistant, there is an external audience beyond the Project Team Leaders and team members that needs to be considered: Project Executing Unit (PEU): plays a key role in the implementation of projects funded by the Bank. The PEU is usually a unit within the government or beneficiary organization responsible for the day-to-day management, overseeing project implementation, ensuring that activities are executed according to the project plan, timeline, and budget, while managing procurement and hiring processes. It handles financial management, tracks expenses, and ensures compliance with MDB guidelines, including transparent procurement and financial reporting. Scope of Work We foresee the following minimum activities for the Knowledge Platform and LessonsGPT AI writer assistant. It is expected that the winner of this contract will work with parallel teams that can evolve the platform, and the Lessons Learned Systematization by-product. The scope of these terms of reference is Release 1 of the product roadmap detailed in the technical specifications annex (Annex 3). These terms of reference are a framework proposal that can be adjusted by the applicant according to its own methodology and approach: Key activities & Description Requirements for product Context and Understanding the institutional context and the constraints and opportunities in terms of technology. coordination Coordination with the bank's Product Design and Tech Side team. Assessment with ITE and KIC. Comprehensive Work Develop a detailed work plan using user story mapping to capture all functionalities and user Plan (User Story interactions across both MVPs. This plan should outline specific user needs, features, and Mapping) workflows, providing a clear visual representation of the user journey. User story mapping will facilitate prioritization of features, ensure alignment with user requirements, and enable the teams to manage development milestones effectively. Sprint plannings Review and prioritize development sprints. Always aligned with the product roadmap. Sprint reviews Per sprint review of the achievements and the working product. Sprint Per sprint retrospectives for continuous improvement. retrospectives Sprint dailies Per day review of the sprint plan from the technology side. Knowledge Platform & Development the components following the design guidelines and the proposed navigation structure LessonsGPT AI writer with an Agile approach. assistant development sprints Accessible The development must conform to the Revised 508 Standards and the W3C development Web Content Accessibility Guidelines (WCAG) 2.0 Level AA. Set up of different The Knowledge Platform and LessonsGPT AI writer assistant must be in two languages: Spanish and languages English, considering it will later be also developed in Portuguese and French. Multilanguage must be configured in the site and a version of every page must be created in every language.  System Architecture The provider must design and document the technical architecture of the platform and its components, including the use of artificial intelligence, API integration, and any necessary infrastructure. Testing Knowledge Platform & LessonsGPT AI writer assistant will be reviewed by the Product Team, tested by users and stakeholders before being approved for publishing. Test will include: Functional testing, Browser and device compatibility testing, Usability testing, Accessibility testing, Performance testing and Stress testing. Deployment Coordinate with the IDB DevOp team to create the CI/CD pipelines. Launching Approved pages and components will be published and monitored. Post-launching corrections may be necessary. Integrations The products are expected to be able to interface with the following bank systems: SAP*, Salesforce*, Denodo (Data Marketplace), MS Office 365, Convergence, ezShare webservices, Data Portal, Evidence Gap Map*, Lessons GPT, Client Portal, SgDelta, Aurora. (*these integrations will require custom development/ configuration for Authentication). Data Integration/ Define API contracts to integrate with IDB internal systems. Implementation of Interfaces mock data API for the data interfaces while development is in course. Secure data integration with Azure API management. Detailed documentation of the developed or integrated APIs, including endpoints, methods, and usage examples, to facilitate future integrations. Interactive Explore the development of interactive components that may include knowledge playlist, to-do lists, Components application shortcuts, custom news, preview widgets, among others. The Offeror may choose to fully develop these components or rely on commercially available third- party components. components in the marketplace that can conform to the design guidelines. Analytics and monitoring Data collection, following metrics prioritized by the product team and creation of a dashboard that dashboards allows monitoring of the platform, including (but not limited to) page and component response time, health alerts on integrations, page and component usage, time on pages, visitor location and other relevant data, while ensuring compliance with privacy guidelines to avoid identification of users on an individual level. Monitoring and final fixes Once the Knowledge Platform and LessonsGPT AI writer assistant and their components are fully deployed, we expect a hyper-care period to ensure optimal performance and functionality. Transfer to support team Products transfer to the support team, of the developed components. Instructions for end-users and the maintenance team, covering both daily use and best practices for support and updates. Documentation of the Detailed documentation is required for the workflow during each development phase, from discovery Work Process and planning to implementation and testing. The product documentation will be done in Confluence. Technical Definitions and A description of the selected technologies, programming languages, frameworks, and any critical Manuals dependencies, with a justification of their choice based on scalability, security, and maintenance. Key Deliverables The following deliverables outline the essential requirements for the development of the two MVPs. This document specifies key deliverables across the implementation of both MVPs, platform evolution, prototyping of new features, and a user-centered approach, all aimed at ensuring high- quality outcomes. The scope emphasizes a modular, scalable architecture, rigorous testing and validation, and a commitment to continuous user feedback and improvement. Key deliverables Description MVP Development Knowledge Platform Development and Integration Design and develop a functional prototype of the Knowledge Management Platform and its components, integrating key capabilities for knowledge search, systematization, and  interaction. The prototype should be compatible with the internal systems and databases of the IDB Group. Quality and User Experience Ensure that the MVP meets the highest standards of quality, performance, and security, facilitating a smooth and intuitive user experience that promotes platform adoption among users. Testing and Validation Conduct thorough testing of the MVP to validate its functionality, stability, and compatibility with existing systems, ensuring it meets the operational and technical requirements of the IDB Group. Integration with Internal and Ensure that the product is compatible with the IDB Group’s internal systems, as well as with External Environments: third-party services and tools that may be necessary for the platform's development or evolution. This includes the capability to interact with external APIs, databases, and cloud services, enabling interoperability and seamless integration. Modular and Scalable Design Build the platform using a modular architecture that allows for the incorporation of changes and updates without requiring a complete redesign. This will facilitate the introduction of new functionalities and modifications with minimal rework, allowing the system to quickly adapt to the emerging needs of the IDB Group. Platform Evolution Workflow Design Establish flexible workflows that enable continuous iteration and expansion of the platform, ensuring that new use cases and functionalities can be integrated as the IDB Group’s needs evolve. Scalability Design the platform architecture to support scalability in terms of users and data volume, as well as the integration of emerging technologies and new AI capabilities. Continuous Update Define an update approach that allows for the incorporation of functionalities in future versions, responding to the needs of different types of users and specific use cases. Prototyping of New Features Development of Mid- and Create detailed prototypes for new use cases and future functionalities, ensuring that these High-Fidelity Prototypes designs accurately represent the expected functionality and usability in future versions of the platform. User Testing Subject prototypes to testing with representative users from different archetypes within the IDB Group, collecting qualitative and quantitative feedback to help refine the design and adjust functionalities based on actual user needs. Feedback-Based Optimization Incorporate observations and suggestions from testing into prototype iterations, ensuring that improvements and new features reflect user expectations and align with the strategic objectives of the platform. User-Centered Approach Intuitive and Accessible Implement a user interface design that is intuitive and easy to navigate, prioritizing the needs Design and preferences of end users. This includes ensuring platform accessibility, according to recognized standards, to make it inclusive and available to people with diverse abilities. Intuitive and Accessible Implement a user interface design that is intuitive and easy to navigate, prioritizing the needs Design and preferences of end users. This includes ensuring platform accessibility, according to recognized standards, to make it inclusive and available to people with diverse abilities. Continuous Feedback and Establish a system that allows users to provide feedback consistently, enabling a quick Improvement response to their needs and iterative improvements based on usage data and comments, promoting continuous platform enhancement. Key deliverables Description MVP Development LessonsGPT AI writer assistant Development and Integration Design and develop a functional prototype of the LessonsGPT AI writer assistant. The prototype should be compatible with the internal systems and databases of the IDB Group. Quality and User Experience Ensure that the MVP meets the highest standards of quality, performance, and security, facilitating a smooth and intuitive user experience that promotes platform adoption among users. Multilanguage (esp, eng, port, French)?  Testing and Validation Conduct thorough testing of the MVP to validate its functionality, stability, and compatibility with existing systems, ensuring it meets the operational and technical requirements of the IDB Group. Integration with Internal and Ensure that the product is compatible with the IDB Group’s internal systems, as well as with External Environments: third-party services and tools that may be necessary for the platform's development or evolution. This includes the capability to interact with external APIs, databases, and cloud services, enabling interoperability and seamless integration. Modular and Scalable Design Build the platform using a modular architecture that allows for the incorporation of changes and updates without requiring a complete redesign. This will facilitate the introduction of new functionalities and modifications with minimal rework, allowing the system to quickly adapt to the emerging needs of the IDB Group. Product Evolution Workflow Design Establish flexible workflows that enable continuous iteration and expansion of the platform, ensuring that new use cases and functionalities can be integrated as the IDB Group’s needs evolve. Scalability Design the architecture to support scalability in terms of users and data volume, as well as the integration of emerging technologies and new AI capabilities. Continuous Update Define an update approach that allows for the incorporation of functionalities in future versions, responding to the needs of different types of users and specific use cases. Prototyping of New Features Development of Mid- and Create detailed prototypes for new use cases and future functionalities, ensuring that these High-Fidelity Prototypes designs accurately represent the expected functionality and usability in future versions of the platform. User Testing Subject prototypes to testing with representative users from different archetypes within the IDB Group, collecting qualitative and quantitative feedback to help refine the design and adjust functionalities based on actual user needs. Feedback-Based Optimization Incorporate observations and suggestions from testing into prototype iterations, ensuring that improvements and new features reflect user expectations and align with the strategic objectives of the platform. User-Centered Approach Intuitive and Accessible Implement a user interface design that is intuitive and easy to navigate, prioritizing the needs Design and preferences of end users. This includes ensuring platform accessibility, according to recognized standards, to make it inclusive and available to people with diverse abilities. To enhance the user experience and make it easier to generate lessons, ensure that the tool includes both text and voice functionality for interaction. Intuitive and Accessible Implement a user interface design that is intuitive and easy to navigate, prioritizing the needs Design and preferences of end users. This includes ensuring platform accessibility, according to recognized standards, to make it inclusive and available to people with diverse abilities. Continuous Feedback and Establish a system that allows users to provide feedback consistently, enabling a quick Improvement response to their needs and iterative improvements based on usage data and comments, promoting continuous platform enhancement. The key deliverables for the project are: Codebase of Custom Components: Complete, documented codebase for all custom modules, including AI-driven features for knowledge search, organization, and interaction. Infrastructure Configuration: Configuration and deployment of the platform’s infrastructure, ensuring compatibility with the IDB Group’s internal systems and any required third-party services. The configuration of the IaC templates (infrastructure as code) as specified in the IDB standards.  Complete Platform Development: Full development of the Knowledge Management Platform in line with the Design Document, covering core functionalities for MVP implementation. Successful Testing and Validation: Comprehensive testing of the platform to verify functionality, stability, and compatibility, ensuring it meets all technical and operational requirements. Report Approval by IDB Project Manager: Submission and approval of all project documentation and progress reports by the IDB Project Manager, confirming compliance with project specifications and deliverables. Key Documentation We expect the following reports on the project: Key report Description Project Plan Work plan detailing activities with their dependencies, timeline, team involved, as well as the objectives of each activity Sprint plans Collaborate with the team's Scrum Master in making detailed descriptions of the developments and changes to be carried out during the sprint Test cases, automated tests Full documentation of test cases and test results for: user acceptance testing, accessibility and Testing results testing both manual and automated. Metrics and dashboard design Metrics proposed including data sources and frequency. Technical Documentation Solution specifications with complete detail of the configurations made to the infrastructure. Accessibility Conformance Create the ACR or ICT testing baseline for accessibility report. Report (ACR) Project Schedule and Milestone The project to develop the MVPS of Knowledge Platform and LessonsGPT AI writer assistant will be carried out over a six-month period, following an agile methodology to ensure flexibility and responsiveness to evolving requirements. This approach enables iterative development and regular feedback loops, allowing the team to deliver functional components incrementally and adapt to the IDB Group’s priorities throughout the project. Key milestones and deliverables will be defined at the start of each sprint, ensuring that progress is aligned with the overall project objectives and timeline. Reporting Requirements To successfully deliver both MVPs, two parallel teams with the following roles will be established. Each team will align closely with the Strategic Alignment Team and the Product Team to meet IDB's functional and technical specifications:  Product Manager: From the vendor's perspective, the Product Manager is responsible for aligning the product’s vision with the client’s needs and business objectives, ensuring successful delivery and measurable impact. DevOps Engineer: Responsible for managing infrastructure, creating and maintaining deployment pipelines, and ensuring smooth continuous integration and delivery. This includes monitoring, scaling, and managing cloud resources to support development and production environments. Software Architect: Designs a modular and scalable architecture that supports both MVPs, ensuring seamless integration with other IDB systems. The architect selects AI technologies and frameworks best suited for the project requirements and maintains compliance with security and privacy protocols. AI Engineer / Machine Learning Engineer: Implements and fine-tunes AI models including natural language processing (NLP) capabilities for intelligent search, recommendations, and insights. Dealing with LLMs, agents and prompting engineering. The AI engineer will also work on embedding and data-processing pipelines tailored to each MVP’s needs. Backend Developer: Develops server-side logic, integrates APIs and databases, and ensures the backend is efficient, secure, and scalable. This includes managing CRUD operations, API connections, and handling secure access to the platform’s data. Frontend Developer: Implements user interfaces based on UX/UI design specifications, ensuring accessibility, usability, and responsiveness across platforms. The frontend developer will create interactive components, ensuring that designs are functional and aligned with IDB’s design standards. QA Engineer / Tester: Conducts comprehensive testing to verify functionality, reliability, and performance of each platform. This role includes unit, integration, and user acceptance testing, ensuring that all components meet IDB’s quality benchmarks. Data Analyst (Optional): Provides insights from user data, helping to optimize user experience and improve AI recommendations through advanced data analysis and visualizations. This role supports data-driven decision-making and assists in refining AI models based on observed patterns. Proposed ceremonies and Agile ways of working This proposal can be modified by the vendor according to its own methodology and approach: Key activities Description Kickoff Team identified, initial coordination meetings. Sprint plannings Review the product backlog to prioritize design & development sprints.  Sprint reviews Per sprint review of the achievements and the working product. Sprint retrospectives Per sprint retrospectives for continuous improvement. Working Sessions Collaborative weekly sessions with Core Team and specific stakeholders Asynchronous daily Per day review of the sprint plan via Teams Sprint stand-up Synchronous daily meeting reviewing of the sprint plan Showcase Monthly update highlighting progress product development with Key Stakeholders Learning Sessions Lead training sessions in agile methodologies for the product team. Acceptance Criteria To ensure the platform meets quality and functional standards, all deliverables will undergo rigorous testing and validation processes. Each component must be fully operational, secure, and compatible with the IDB Group’s internal systems, as well as adhere to industry best practices for performance and scalability. Acceptance criteria will include successful completion of user testing for both frontend and backend functionalities, with any identified issues resolved before the final handoff. Additionally, the platform must demonstrate stable integration with external APIs and databases, showcasing its ability to support a seamless user experience. Furthermore, documentation will be essential for each phase of the project. The team must provide comprehensive technical documentation, covering the architecture, configuration details, and maintenance protocols. All user-facing components should meet accessibility standards and be easy to navigate, while backend elements should support scalability and ease of modification. The project’s acceptance will depend on the delivery of these documents, along with approval from the IDB Project Manager, confirming that all aspects of the platform align with the IDB Group’s strategic objectives and technical requirements. Eligibility This call is open to registered companies and organizations that meet the following criteria: The organization must be registered or incorporated in one of the 48 member countries of the IDB Group. Demonstrated experience in Artificial Intelligence (AI) and Large Language Models (LLMs). Ability to interact and connect with various stakeholders at different levels. Strong proficiency in English and/or Spanish. Documentation can be delivered in either language. Demonstrated capabilities in Agile Methodology and User-Centered Design.  2. Payment Schedule Percentage Deliverables 15% Project Initiation and Initial documentation, detailed project plan, and technical and architectural Planning specifications. 20% MVP Development Initial MVP version with prioritized functionalities (such as search, basic knowledge and Initial Testing interaction, and initial AI integrations). 15% Systems and API Integration of the MVP with internal and external systems, including APIs and Integration databases. 15% Quality and Security Results of comprehensive quality and security testing, with adjustments made based on Testing feedback. 20% Final MVP Delivery Delivery of the complete MVP version, including all documented modules and user, API, and Documentation and maintenance manuals. 15% Project Closure and Final adjustments, initial post-launch support, and a general compliance review. Final Acceptance Supervision Jose Freixinios (ITE/ITO), Roberto Almeida (ITE/IPP), Lourdes Gallardo (ITE/DT), Mariana Gutierrez (ITE/TECHLAB) 3. Annex Annex C.1 Knowledge Platform Design and Wireframes Document.  Annex C.2 Reference Architecture