The Organization
The Alliance of Bioversity International and the International Center for Tropical Agriculture (CIAT) delivers research-based solutions that harness agricultural biodiversity and sustainably transform food systems to improve people's lives. Alliance solutions address the global crises of malnutrition, climate change, biodiversity loss, and environmental degradation.
With novel partnerships, the Alliance generates evidence and mainstreams innovations to transform food systems and landscapes so that they sustain the planet, drive prosperity, and nourish people in a climate crisis.
The Alliance is part of CGIAR, a global research partnership for a food-secure future.
About the Consultancy
The Performance, Innovation and Strategic Analysis for Impact team within the Alliance seeks to enhance its capacity to conduct in-depth political economy and policy analysis (PEPA). To this end, we are looking for an AI consultant to contribute to developing a cutting-edge simulation tool that systematically facilitates this process. This tool will leverage large language models (LLMs) and retrieval augmentation generation techniques to aid the analysis process, ensuring that the findings are effectively synthesized and presented to policymakers.
Objectives
The primary objectives of this consultancy are to:
Design and develop a user-friendly, web-based simulation tool that automates political economy and policy analysis.
Utilize open-source LLMs and AI tools, where appropriate, to create an efficient and cost-effective solution.
Implement multi-stage AI techniques such as retrieval augmentation generation (RAG) to tailor the LLMs for content using specific organization data.
Deploy the model via an API so other applications and services can use it, primarily the web-based simulation tool.
Document the technical and user platforms and processes to ensure future adaptability and maintainability.
Scope of Work
The AI consultant will:
Conduct a user-centered assessment to understand the requirements and existing ecosystem for the simulation tool.
Design the architecture of the simulation tool incorporating the PEPA six-step process, using LLMS and RAG or similar techniques.
Develop the AI simulation tool using Python programming language, open-source libraries, and frameworks, and the front end using Next.js and open-source JavaScript libraries.
Integrate a PISA-supplied data store with the simulation tool to focus the knowledge base of the LLMs on known documentation.
Test and refine the simulation tool to ensure accuracy, efficiency, and user-friendliness and provide training and comprehensive documentation to PISA staff.
Deliverables
A fully functional simulation tool for PEPA, complete with a user guide, source code, and documentation.
A comprehensive report detailing the development process, including methodologies, data structures,