Mid-Level AI Engineer (Consultant)

About the role

The Individual Contractor will design, develop, and deploy production‑grade agentic AI systems and LLM‑powered solutions for the Arab Development Portal and broader ESCWA innovation initiatives. This short‑term consultancy (3 months) supports ESCWA’s Decision‑Support and Data Science Division in delivering advanced analytics and AI capabilities across the region.

Key responsibilities

• Architect and implement multi‑agent systems using frameworks such as LangGraph, AutoGen or CrewAI, including retrieval‑augmented generation (RAG) pipelines and tool‑use capabilities.
• Integrate, benchmark, and fine‑tune state‑of‑the‑art LLMs (OpenAI, Anthropic, Mistral, HuggingFace, Ollama) for Arabic and multilingual tasks such as extraction, summarisation and question answering.
• Design and expose RESTful and streaming APIs, containerise services with Docker, and maintain documentation, versioning and model cards.
• Collaborate with data engineers, ML engineers and domain experts to embed AI components into end‑to‑end data products and produce technical reports and demonstrations.

Required profile

• Bachelor’s degree in Computer Science, AI, Data Science or related field (Master’s preferred); proof of degree required.
• Minimum 5 years of professional experience in AI/ML or software engineering with a strong focus on AI.
• Fluency in English (spoken, written, read, understood) and ability to work in a UN environment.

Required skills

• Python programming.
• Multi‑agent frameworks (LangGraph, AutoGen, CrewAI or equivalents).
• LLM APIs (OpenAI, Anthropic, Mistral, HuggingFace, Ollama) and prompt engineering.
• RAG architectures and vector databases (FAISS, Weaviate, Qdrant).
• Docker containerisation and RESTful/streaming API development.
• Model fine‑tuning techniques (LoRA, QLoRA) and evaluation methodologies.
• Knowledge of Arabic NLP and multilingual model handling.