AI/Large Language Model Consultant – DRC

Background

SoCha, LLC is contracted to provide technical and advisory services to the United States Agency for International Development in the Democratic Republic of Congo (USAID/DRC) on the 5-year Mission Strategic Support Program (MSSP). The purpose of MSSP is to provide these services in planning, designing, conducting, disseminating, and learning from monitoring and evaluation of the Mission’s portfolios, assisting the mission with strategy and activity-level performance monitoring, evaluation, and learning, including conducting data gathering and verification, data visualization (including Geographic Information Systems), data quality assessments, impact and performance evaluations, assessments, organizational learning, and collaboration among USAID’s implementing partners. The activity will also implement a capacity-building agenda for the Mission and implementing partners, developing a baseline of monitoring and evaluation knowledge within the Mission and tailoring a curriculum to provide skills development to Mission staff.

USAID/MSSP is set to develop a Retrieval-Augmented Generation system to improve access and management of extensive document databases via Microsoft SharePoint and other platforms. This project will utilize cutting-edge AI technologies to streamline information retrieval and enhance user interaction with vast data repositories. An experienced AI/Large Language Model Expert Consultant is sought to spearhead this initiative.

Qualified consultants are eligible to apply, regardless of their nationality or place of residence.

Roles and Responsibilities

In collaboration with the Data Management Specialist and the IT team, the selected candidate will:

• Recommend a technology stack based on the requirements.
• Provide a detailed cost proposal and implementation plan.
• Design an integration with SharePoint for optimized document retrieval and data management.
• Develop a chat interface using Large Language Models for conversational document querying.
• Build a model-agnostic data pipeline for API integration, data preprocessing, and dynamic updates.
• Set performance metrics and conduct validations to ensure relevance and accuracy.
• Iteratively enhance the system based on user feedback to improve usability and data precision.
• Facilitate the translation of complex AI processes into user-friendly functionalities.
• Document workflows and system architecture and conduct comprehensive end-user training.

Minimum Qualifications and Expertise

• Master’s degree in computer science, AI, or related fields.
• More than four (4+) years of professional experience in AI and machine learning, with proven ability in managing large document databases and vector databases.
• Proven ability to develop implementation plans, manage timelines, and communicate with stakeholders.
• Extensive Python experience with the AI libraries such as Transformers, PyTorch LangChain, Pandas, spaCy, and the API client for OpenAI.
• Experience designing user-friendly systems for both technical and non-technical users.
• Proven track record involving document retrieval, conversational AI or knowledge management systems.
• Experience with Azure Cloud services would be an advantage.
• Experience with SharePoint and Microsoft Graph API would be an advantage.

Location

Kinshasa, Democratic Republic of Congo, or Remote.

Duration

The assessment is projected to last for nine weeks.