Software Engineering Intern, AI Tools - Fall 2026

NVIDIA NVIDIA · Semiconductors · Santa Clara, CA

NVIDIA is seeking a Software Engineering Intern to join its AI Tools and Infrastructure team, focusing on Agentic AI. The intern will work with LLMs and orchestration frameworks to design, build, and deploy intelligent agents and AI tools, gaining exposure to the full lifecycle of agent development from conceptualization to deployment.

What you'd actually do

  1. Design and rapidly prototype LLM-powered agents that exhibit planning, reasoning, memory, and tool-use capabilities to automate complex, multi-step tasks.
  2. Craft, refine, and test high-performance prompts, instructions, and system messages to guide agent behavior and optimize task completion accuracy.
  3. Work with agent orchestration frameworks to connect agents with internal and external APIs, databases, and knowledge bases (e.g., Vector DBs for RAG) to enable real-time actions.
  4. Develop metrics and meticulous testing workflows to benchmark agent performance, reliability, and safety across various use cases.
  5. Maintain clear user documentation of agent architectures, workflows, and experimental results.

Skills

Required

  • Pursuing a BS, MS or PhD with a focus on Computer Science, Computer Engineering, or a related field
  • Conceptual understanding of LLMs, Generative AI, and core Machine Learning concepts.
  • Demonstrated passion or basic project experience with Agentic AI concepts (e.g., RAG, MCP, multi-step planning, or using frameworks like LangChain/LlamaIndex).
  • Familiarity with software development guidelines, including version control (Git/GitHub) and basic testing/debugging.
  • Problem solving and analytical thinking, with strong algorithmic, design and debugging skills
  • Proficient in at least one programming language such as C, C++, C#, Java, or Python (Python preferred)
  • Experience with writing and consuming APIs, and working with web application frameworks
  • Strong verbal and written communication skills

Nice to have

  • Curiosity about different areas of software engineering and an eagerness to leave your comfort zone
  • An eagerness for learning and working on the cutting edge of technology

What the JD emphasized

  • Agentic AI
  • LLM-powered agents
  • orchestration frameworks
  • autonomous agents
  • agent development
  • multi-step tasks
  • tool-use capabilities
  • prompt engineering
  • Vector DBs for RAG
  • agent performance, reliability, and safety

Other signals

  • LLM-powered agents
  • orchestration frameworks
  • autonomous agents
  • generative and autonomous systems
  • agent development