AI Application Engineer (Internship） 工作, 米连集团 Federal Territory 公司招聘中 - Ricebowl

Position : AI Application Engineer (Agent / Game Creation Platform) — Internship

Role Overview

Contribute to the R&D of the Agent system for an AI game creation platform. Building on the company's existing game editor and runtime, enable users to create and modify games through natural language. The role focuses on the Agent execution loop, tool integration, art asset automation, and reliability engineering. It does not require training or algorithmic development of underlying generative models. This position starts as an internship, and we strongly welcome applications from current Master's and PhD students.

Responsibilities

• Contribute to the R&D of the Agent system for the AI game creation platform. Building on the company's existing game editor and runtime, enable users to create and modify games through natural language.
• Contribute to the development of the Agent execution loop for game creation, including intent understanding, task planning, tool invocation, context and execution-state management, result validation, and error recovery, supporting multi-step task completion over extended interactions.
• Collaborate with the editor and game R&D teams to encapsulate existing capabilities—such as maps, characters, skills, and gameplay rules—and integrate them into the Agent, supporting flexible configuration and composition.
• Contribute to the game art asset generation and processing pipeline. By invoking tools or services for image generation, image editing, background removal, motion generation, video frame extraction, and sprite-sheet processing, enable automation from user requirements to usable game assets.
• Collaborate with the art and asset-system R&D teams to complete resource-generation task orchestration, asynchronous status tracking, result validation, failure retry, and asset ingestion, and apply assets in-game through the editor.
• Contribute to operational reliability. Handle duplicate requests, execution failures, and related issues; support task recovery, operation tracking, and undo.
• Build Agent test cases and evaluation processes. Through experiments, log analysis, and failure-case reviews, continuously optimize task success rate, interaction experience, response speed, and invocation cost.

Requirements

• Master's degree or above in Computer Science, Software Engineering, Artificial Intelligence, Automation, Mathematics, or a related field. Current Master's/PhD students and fresh graduates are all welcome.
• Solid programming and computer-science fundamentals. Proficient in at least one of Python, TypeScript, or Go, with the ability to independently complete module development, debugging, and testing.
• Hands-on project experience with Agents, from research, internships, open source, or personal projects. Practical involvement in a multi-step task flow—"understanding user requirements, decomposing tasks, invoking tools or APIs, and continuing based on execution results"—with the ability to show code or running results and explain your own contributions, problems encountered, and how they were solved. Project experience should not be limited to Q&A applications or simple model API calls.
• Understanding of structured output from large models, tool invocation, context management, and task planning; awareness of model capability boundaries; ability to reasonably divide responsibilities between model inference and deterministic program logic.
• Basic software engineering and system-integration skills. Familiarity with HTTP APIs, databases, asynchronous tasks, and exception handling; ability to integrate existing business capabilities into an Agent.
• Ability to design experiments and analyze problems: construct test cases, analyze failure causes, and validate optimization effects through controlled experiments.
• Proactive learning, communication, collaboration, and execution skills. Willingness to deeply understand game-creation scenarios and translate technical solutions into usable product features.

Preferred Qualifications

• Agent project experience with natural-language-driven editors, AI programming, low-code builders, design tools, or other complex business operations.
• Project experience with image generation and editing, image processing, video or motion generation, or sprite-sheet processing.
• Research experience in Agent planning, memory, tool use, or evaluation.
• AI application R&D internship experience, open-source contributions, or personal projects used by real users.
• Experience in game development, configuration-driven systems, or automated testing.