AI Engineer Intern

About zCloak.AI zCloak.AI helps enterprises transform business operations through production-ready AI workflows. We build an AI agent work platform that connects enterprise data, tools, and teams with secure, governed AI execution. The Role We are looking for an AI Engineer Intern to help design, build, test, and improve the AI systems that power zCloak.AI. You will work on AI agents, retrieval systems, tool-use infrastructure, agent memory, workflow orchestration, evaluation, and model integration. Your work will span the full AI application stack — from models and prompts to retrieval, tools, runtime systems, observability, and production reliability. This is a hands-on engineering internship with meaningful ownership of real AI product and engineering work. You will work closely with product, platform, and deployment teams to turn emerging AI capabilities into dependable product features that can operate inside real enterprise workflows. You should be comfortable experimenting rapidly, measuring system behaviour, debugging technical issues, and helping turn successful prototypes into reliable and maintainable product features. What You Will Do Help design, build, test, and improve AI agents and agentic systems used in production workflows. Develop RAG pipelines, retrieval systems, document-processing pipelines, and knowledge-grounding mechanisms. Build and improve tool-use systems, agent memory, workflow orchestration, approval flows, and human-in-the-loop mechanisms. Integrate and evaluate foundation models from multiple providers and help determine the appropriate model, prompting, routing, and execution strategy for different tasks. Develop structured-output, function-calling, and multi-step reasoning workflows for enterprise use cases. Build evaluation frameworks to measure task completion, accuracy, reliability, latency, and cost. Design and implement guardrails, validation mechanisms, fallback strategies, and failure-recovery logic for AI systems. Develop tracing, logging, monitoring, and observability capabilities for agent execution. Investigate and resolve failures across models, prompts, retrieval, tools, data pipelines, application code, and infrastructure. Improve model and agent performance through prompt optimisation, retrieval improvements, model selection, context management, and system-level engineering. Contribute to reusable AI components, internal libraries, SDKs, and platform capabilities that support multiple products and customer deployments. Work with product and engineering teams to translate product requirements into practical AI system designs and implementations. Evaluate new models, agent frameworks, research developments, and AI infrastructure, and identify where they can create practical product improvements. Contribute to technical architecture, engineering standards, testing practices, and system documentation. Support debugging, testing, and incident investigation where AI system behaviour is involved. Minimum Qualifications Currently pursuing or recently completed a Bachelor’s or Master’s degree in Computer Science, Software Engineering, Artificial Intelligence, Machine Learning, Data Science, or a related technical field, or equivalent practical experience. Candidates from quantitative or analytics programmes with strong hands-on AI engineering experience are also encouraged to apply. Strong programming ability in Python. Solid software engineering fundamentals, including APIs, databases, testing, debugging, and version control. Hands-on experience building applications with LLMs, AI agents, tool calling, or workflow orchestration. Hands-on experience or substantial project work involving RAG systems, embedding models, vector databases, retrieval techniques, or document-processing tools. Understanding of modern LLM application patterns, including prompting, structured outputs, function calling, context management, and model evaluation. Experience integrating APIs, databases, model providers, or other external tools through internships, projects, research, or open-source work. Experience working in Linux-based development or deployment environments. Ability to investigate ambiguous technical problems and systematically identify root causes. Ability to take an AI feature or project from initial experimentation through implementation, testing, and iteration. Clear written and verbal communication skills in English and Chinese. Preferred Qualifications Internship, research, open-source, competition, or project experience building AI applications, agent systems, machine learning systems, or developer tools. Experience with one or more agent frameworks or SDKs, such as LangGraph, OpenAI Agents SDK, Google ADK, CrewAI, OpenClaw, Hermes, or equivalent systems. Familiarity with MCP, agent memory, tracing, observability, evaluation, and prompt or model versioning. Experience designing multi-agent systems or long-running workflow orchestration. Experience building model evaluation pipelines, test datasets, benchmarks, or automated regression testing for AI systems. Familiarity with model routing, caching, context management, inference optimisation, and AI application cost optimisation. Experience with vector databases and retrieval infrastructure such as pgvector, Qdrant, Milvus, Pinecone, Weaviate, or similar systems. Familiarity with AWS, Google Cloud, Azure, containers, CI/CD, and infrastructure automation. Understanding of enterprise security concepts, including identity, permissions, secrets management, audit logs, and data governance. Experience working with document-heavy or workflow-heavy enterprise applications. Experience working in a startup or another fast-moving engineering environment. Contributions to open-source AI projects, relevant research, technical publications, or substantial deployed AI projects. We welcome students with strong hands-on AI engineering experience gained through internships, research, open-source contributions, competitions, or substantial personal or academic projects. We care more about demonstrated technical ability, curiosity, and ownership than years of professional experience. What Success Looks Like A successful AI Engineer Intern can take an AI problem from an initial idea or prototype to a well-tested, working implementation, with guidance where needed. You will be able to identify why an AI system fails, determine whether the problem comes from the model, prompt, retrieval, tools, data, orchestration, or surrounding application logic, and implement practical improvements. You will contribute to systems that become progressively more accurate, reliable, observable, efficient, and reusable, while developing the engineering judgment needed to build dependable enterprise AI products. Apply If you want to build AI systems that operate inside real enterprise workflows, we would like to hear from you. Please send your CV and, where available, your GitHub profile, portfolio, technical writing, research, or examples of relevant projects to: yuyang@zcloak.network Email subject Application – AI Engineer Intern – [Full Name] – [School] – [Internship Availability / Earliest Start Date] Learn more about zCloak.AI: www.zcloak.ai