Research Consultant

Consultant IT & Telecom Information Management

Overview

Support AI-driven generation of the CCA for Kuwait, enhancing insights and analysis.

You have:

• A Bachelor's degree in Data Science, Artificial Intelligence, Machine Learning, Statistics, Economics or related field is required.
• A minimum of 5 years of professional work experience in the application of artificial intelligence, data science, or digital tools for research, analytics, or programme development is required.
• Proven experience in AI-driven text processing and document analysis is required.
• Proficiency in GPT-based LLMs, LangChain, or similar frameworks is required.
• Proficiency in vector databases (e.g., FAISS, ChromaDB) is required.
• Strong understanding of retrieval-augmented generation (RAG) and prompt engineering is desirable.
• Technical background in data science, machine learning, or NLP is desirable.
• Familiarity with sustainable development and UN planning frameworks is desirable.
• Strong communication and independent problem-solving skills is desirable.
• Fluency in English is required.
• Knowledge of Arabic is desirable.

Contract

This is a Consultancy contract. More about Consultancy contracts.

Result of Service

Under the guidance of the RC and supervision of the Senior Economist, the consultant will be responsible for undertaking all necessary activities to support the successful completion of the CCA process. Specifically, the consultant will: • Enhance the Retrieval-Augmented Generation (RAG) Pipeline: o Refine existing AI-driven pipelines to improve semantic document indexing and response generation. o Ensure compatibility and responsiveness of the system using orchestration framework (LangChain) and a modern LLM (LLaMA3 or equivalent). • Support Document Ingestion and Vector Embedding: o Preprocess and convert thematic documents (PDF, DOCX) into machine-readable formats. o Structure and embed documents into a vector database (e.g., FAISS, ChromaDB). • Optimize Prompt Engineering and Output Structuring: o Develop and test prompts to generate structured outputs aligned with the CA summary template. o Ensure coverage of sections including root causes, affected populations, interlinkages, SDG alignment, financial dimensions, and opportunities for change. • Evaluate and Refine Model Outputs: o Compare AI-generated summaries with manual inputs to enhance relevance and accuracy. o Ensure model outputs reflect UN analytical language and maintain methodological consistency. • Support Integration into Final CA Outputs: o Assist in preparing final, SAT-validated summaries for each development challenge. o Propose ways to scale or replicate the approach for other UN analytical processes, and explore ways to create a Chat Bot able to communicate with the user.

Work Location

Remote

Expected duration

2 Months

Duties and Responsibilities

Background: The UN Resident Coordinator Office (RCO) leads the coordination and strategic alignment of the UN system’s support to Kuwait in line with the Sustainable Development Goals (SDGs), UN mandates, and Kuwait Vision 2035. The RCO forges partnerships with the Government of Kuwait and serves as the Secretariat for inter-agency coordination among 11 Resident and 15 Non-Resident UN entities. The RCO is launching the 2025 Common Country Analysis (CCA), the UN’s independent and forward-looking assessment of Kuwait’s development context. This year’s CCA aims to be a transformative, data-driven and highly visual product that informs both internal programming and external engagement. It will move beyond traditional reporting to provide actionable insights on Kuwait’s post-oil economic transition, regional instability and national development goals. Scope of work The Consultant will provide support for the application of AI for the generation of the CCA, for the first stage of the CCA generation (desk review), including supplementing sections that are missing or need to be strengthened again with the use of AI, suggesting and implementing additional analysis as needed, and as much as possible, including a disaggregated assessment and analysis in the AI desk review of groups at higher risk of vulnerability (women, youth, elderly, persons with disability, minorities, migrants and refugees, others) and subnational assessment and analysis.

Qualifications/special skills

A Bachelor's degree in Data Science, Artificial Intelligence, Machine Learning, Statistics, Economics or related field is required. All candidates must submit a copy of the required educational degree. Incomplete applications will not be reviewed. A minimum of 5 years of professional work experience in the application of artificial intelligence, data science, or digital tools for research, analytics, or programme development is required. Proven experience in AI-driven text processing and document analysis is required. Proficiency in GPT-based LLMs, LangChain, or similar frameworks, and vector databases (e.g., FAISS, ChromaDB) is required. Strong understanding of retrieval-augmented generation (RAG) and prompt engineering is desirable. Technical background in data science, machine learning, or NLP is desirable. Familiarity with sustainable development and UN planning frameworks is desirable. Strong communication and independent problem-solving skills is desirable.

Languages

English and French are the working languages of the United Nations Secretariat; and Arabic is a working language of ESCWA. For this position, fluency in English is required. Knowledge of Arabic is desirable. Note: “Fluency” equals a rating of ‘fluent’ in all four areas (speak, read, write, and understand) and “Knowledge of” equals a rating of ‘confident’ in two of the four areas.

Additional Information

Not available.

No Fee

THE UNITED NATIONS DOES NOT CHARGE A FEE AT ANY STAGE OF THE RECRUITMENT PROCESS (APPLICATION, INTERVIEW MEETING, PROCESSING, OR TRAINING). THE UNITED NATIONS DOES NOT CONCERN ITSELF WITH INFORMATION ON APPLICANTS’ BANK ACCOUNTS.

Apply now

Potential interview questions

Added 7 months ago - Updated 7 months ago - Source: careers.un.org