Agentic AI Research Intern (home-Based)

Organizational Context

The United Nations Development Programme (UNDP) is the lead UN agency for international development, working in 170 countries to eradicate poverty and reduce inequality. The Istanbul International Center for Private Sector in Development (ICPSD), a Global Policy Center, focuses on private sector engagement and innovative solutions for sustainable development, with a specific pillar dedicated to AI and digital transformation.

Job Purpose

This internship offers a unique opportunity to support the ICPSD SDG AI Lab's work on Agentic AI, contributing to the research and development of digital tools for sustainable development. The intern will gain hands-on experience in developing Agentic AI solutions, including autonomous agents, multi-agent systems, and Large Language Models (LLMs), to address global development challenges like poverty, inequality, and digital inclusion. The role provides practical exposure to UNDP's standards and workflows within an international development context, fostering collaboration with a multicultural team and contributing to UNDP's mission.

Responsibilities

The Agentic AI Research Intern will support the ICPSD Digital Transformation, Skills and AI Portfolio's research activities. Key duties include conducting research on Agentic AI applications for global development challenges, drafting analytical briefs on project findings, and supporting the development, implementation, and testing of Agentic AI solutions. This involves developing and optimizing multi-agent workflows using frameworks like LangGraph or CrewAI, building proof-of-concept agentic systems with tools such as LangChain or Microsoft Copilot Studio, and creating tools and APIs for agent interaction with development databases. The intern will also support the implementation of RAG systems and vector databases, and provide administrative support as needed.

Work Experience

Required experience includes proficiency in Python programming for agent workflows and APIs, strong knowledge of prompt engineering for LLMs, and experience implementing agentic AI systems using frameworks like LangChain or CrewAI. Desirable experience includes familiarity with various LLMs, RAG systems, vector databases, and evaluation frameworks. Experience with Microsoft Copilot Studio or Power Platform is an asset.

Skills

Proficiency in Python programming, API development, prompt engineering, experience with agentic AI frameworks (LangChain, LangGraph, CrewAI), research, writing, and analytical skills, strong communication and interpersonal skills, ability to work independently and collaboratively, and commitment to UNDP's mission.

Explore related opportunities