AI Agent Engineering Project Intern - Procurement

About the role

ByteDance is upgrading its traditional procurement system into an AI-native procurement platform, focusing on building Procurement Agents and the underlying System AI Harness. As a Project Intern, you will work alongside experienced engineers on short-term projects, with hands-on work from day one and mentorship. The internship starts in 2026 and is open to candidates currently pursuing a Bachelor's or Master's degree.

Key responsibilities

• Support the construction and exploration of Procurement Agents and the System AI Harness, contributing to feature design, implementation and validation under your mentor's guidance.
• Develop and maintain services, business logic and data flows that power AI-native procurement, with attention to code quality and reliability.
• Explore AI Agent capabilities such as planning, tool use, memory and multi-step reasoning, and help turn them into working features.
• Build quick prototypes, experiment with emerging AI techniques and help turn promising ideas into practical product capabilities.
• Collaborate with product managers, designers and engineers to deliver complete solutions.

Required profile

• Currently pursuing a Bachelor's or Master's in Computer Science, Software Engineering, AI, Data Science, Electronic/Information Engineering, Mathematics or a related technical discipline.
• Able to start immediately, ideally within one week, and to intern continuously for at least 3 months, at least 3 days per week.
• Solid foundation in data structures, algorithms and software engineering fundamentals.
• Genuine curiosity and passion for frontier AI technologies; prior professional experience is not required.

Required skills

• Proficiency in at least one programming language (Golang, Java or Python).
• Preferred: coursework or projects in AI / machine learning / large language models, exposure to LLM-based systems, prompt engineering, RAG, tool/function calling or agent orchestration, and achievements in programming competitions or open-source contributions.