Artificial Intelligence Engineer Consultant - PISA at Bioversity International

• 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, and AI models used, as well as a guide for future maintenance and updates. 
• Training materials and sessions for the organization's staff on how to use and maintain the simulation tool. 

Requirements

• Advanced degree in Computer Science, Artificial Intelligence, or a related field. Demonstrated experience in AI and machine learning, especially in Natural Language Processing. 
• Proficiency in Python, JavaScript, and open-source AI tools, models, and libraries. 
• Strong analytical, problem-solving, and communication skills and ability to work independently and collaborate with a multidisciplinary team.  
• Background or experience working with Economic or Policy domains will be an added advantage.