AI Automation and NLP Engineering Intern

Organizational Context

The Environment and Development Division of the UN Secretariat's Economic and Social Commission for Asia and the Pacific is seeking an intern to support an internal automation platform. This platform leverages local large-language-model inference to transform UN meeting media into structured reporting and analytical outputs. The role is remote or based in Bangkok, Thailand, for 4-6 months.

Job Purpose

The purpose of this internship is to support the development and enhancement of an internal automation platform within the Environment and Development Division. The intern will contribute to transforming UN meeting media into structured, verified reporting and analytical outputs by utilizing local large-language-model inference. This involves working with a mature Python codebase, focusing on software quality assurance, evaluating models and prompts, and developing tooling for ongoing automation initiatives. The role aims to improve the efficiency and accuracy of data extraction and reporting from multimedia sources, contributing to the division's analytical capabilities and operational effectiveness. The intern will gain practical experience in AI, NLP, and software engineering within a UN context.

Responsibilities

The intern will contribute to a mature Python codebase, focusing on enhancing software quality assurance and developing automation tooling. Key responsibilities include authoring and extending automated tests (unit and mocked-LLM tests) to increase coverage and harden data-extraction paths. The intern will design and execute structured model and prompt evaluations, such as per-language machine-translation selection and speech-recognition error detection, creating measurable results. Additionally, they will build and extend deterministic extraction tooling, ensuring language models do not supply structured identifiers. The role also involves scaffolding new modules and utilities within the existing architecture, improving documentation, refactoring code, and enhancing developer tooling to support ongoing automation initiatives.

Work Experience

While not explicitly stated as a strict requirement, the role involves contributing to a mature Python codebase. Experience with Python, testing frameworks (like pytest), and version control (like git) is required. Familiarity with natural language processing, large language model prompting and evaluation, and running local LLMs is also required or considered an asset. Experience designing evaluation harnesses or working with structured JSON model outputs is desirable.

Skills

Strong Python proficiency; Experience with testing frameworks (pytest); Experience with version control (git); Foundational understanding of Natural Language Processing (NLP); Understanding of large-language-model prompting and evaluation; Familiarity with running/calling local LLMs (e.g., Ollama, Hugging Face); Experience designing evaluation harnesses; Experience with structured (JSON) model outputs; Clean, documented, and reproducible coding practices; Familiarity with responsible AI tool usage.

Explore related opportunities