NVIDIASoftware Engineering Intern, Multi-Agent Systems - 2027
NVIDIA is hiring a Software Engineering Intern, Multi-Agent Systems - 2027 in Shanghai, China. Level rates it ; you can apply on Level.
AI in this role
NVIDIA is seeking a highly motivated and creative Software Engineer Intern to join our SW Infra Team. As an intern, You will play a key role in helping build multi-agent systems and enterprise agent harnesses—the runtime and tooling that enable AI agents to use tools, collaborate, and complete engineering tasks reliably.
This role is ideal for someone interested in turning LLM capabilities into practical software systems. You will work with engineers and researchers to develop agent infrastructure for workflows such as automated regression testing, code review and generation, engineering knowledge retrieval, and Autonomous Vehicle (AV) road event triage.
What You'll Be Doing:
- Design and implement agent harnesses that support tool execution, context and state management, and coordination across multiple agents.
- Build integrations with enterprise tools, code repositories, data platforms, and engineering services to help agents complete real-world development tasks.
- Develop reliable execution capabilities, including sandboxing, access controls, error recovery, and human review and intervention.
- Build evaluation and observability tools to measure agent task success, trace execution, replay trajectories, and improve performance, latency, and cost.
- Collaborate with engineers to prototype agent workflows, validate them against realistic tasks, and improve them through experimentation and feedback.
What We Need to See:
- Currently pursuing a master’s, or PhD in Computer Science, Electrical Engineering, or a related field.
- Programming experience in at least one language, such as Python, C, C++, Rust, or TypeScript, and an understanding of software engineering fundamentals, including data structures, APIs, testing, and debugging.
- Basic familiarity with LLM APIs and an interest in agent systems. Exposure to concepts such as prompting, agent loops, skills, subagents, Model Context Protocol (MCP), tool use, or agent memory through coursework or personal projects is a plus.
- Hands-on software development experience through coursework, research, internships, open-source contributions, or personal projects.
- Interest in multi-agent systems, with the ability to investigate problems, experiment, and communicate technical findings clearly.
- Regular use of agentic coding tools such as Codex, Claude Code, Cursor, or similar products, with curiosity about how they work and an interest in exploring their underlying implementation.
Ways to Stand Out from the crowd:
- Experience building an agent harness or multi-agent application, including task delegation, shared state, memory, or coordination between agents.
- Experience testing an agent on sample tasks and comparing results to identify improvements.
- Basic familiarity with Linux, Git, containers, or cloud platforms.
- Interest in secure agent execution, including permissions, sandboxing, and human review.
- Open-source contributions or small projects using AI agents to automate coding, testing, or data analysis.
Widely considered to be one of the technology world’s most desirable employers, NVIDIA offers highly competitive salaries and a comprehensive benefits package. As you plan your future, see what we can offer to you and your family www.nvidiabenefits.com/
#deeplearningHow we rate this
Software Engineering Intern, Multi-Agent Systems - 2027 at NVIDIA rates 98 out of 100 for how much of the daily work is AI. That makes it Builds AI (AI Level 4 of 4). The level is about AI in the job, not seniority.
Builds AI. The job is building AI systems.
- ●●●● Builds AI80 to 100
- ●●●○ Works on AI60 to 79
- ●●○○ Uses AI40 to 59
- ●○○○ Little AI0 to 39
Levels come from how often the tools, models and workflows of the role are named in the posting itself. Open the description and count.
Prepare for this job
Built from this posting and from similar jobs on Level.
Skills and AI tools this role asks for
Questions you could be asked
- How do you decide when an AI agent can act on its own versus asking for approval first?
- Walk me through how you've used Claude in your day-to-day work.
- What are the limits of Cursor that you've run into, and how did you work around them?
- What's a project where you used Claude Code hands-on?
- Walk me through how you've used Codex in your day-to-day work.
Adapt your resume
- List these exact terms on your resume: AI agents, Claude, Cursor, Claude Code, and Codex. An applicant tracking system matches the wording, not the idea.
- Attach one line of real, concrete experience to at least one of them. A tool named with nothing behind it rarely survives a human read.
- Lead with what you built, trained or shipped. This role is judged on the AI system itself, not the tools around it.
Want an expert to read your CV for this job?
Free. Send your CV and the role you want next. We reply by email within 2 to 4 business days.
Get new software engineering jobs (Builds AI ●●●●) by email
One email a week with the new software engineering jobs (Builds AI ●●●●), each rated for how much AI is in the work. No recruiter spam, unsubscribe in one click.
Free. One email a week. Unsubscribe in one click.
Similar roles
Software Engineering roles that build AI, at other companies.
What kind of AI work fits you?
Answer 12 practical questions in about three minutes. Get a simple profile, the work it points to, and live roles to explore next.
Find my next step