NVIDIA's Deep Learning Safety team is looking for a Software Engineer to build intelligent, autonomous software for the next generation of accelerated computing, specifically designing and implementing AI-driven agents that autonomously interact with APIs, tools, and environments to accomplish complex goals.
Requirements
- Strong programming experience in Python and C++ (or Go/Rust), including experience with async runtimes and API orchestration.
- Experience building applications or systems that incorporate LLMs or foundation models in an interactive or autonomous fashion.
- Familiarity with tools and frameworks for building AI agents (e.g., LangChain, CrewAI).
- Proven systems thinking and ability to design for safety, observability, and robustness in autonomous execution.
- Demonstrated work on agent-based systems, autonomous workflows, or intelligent software automation.
- Experience integrating AI with GPU computing workflows (e.g., CUDA, PyTorch, Triton, or compiler toolchains).
- Knowledge of planning algorithms or program synthesis using LLMs.
Responsibilities
- Design and develop intelligent software agents capable of planning, reasoning, and acting across tools, APIs, and compute environments.
- Integrate agentic AI systems with NVIDIA platforms, including GPU computing frameworks and runtime engines.
- Work with LLMs and agentic frameworks to build robust, reactive, and proactive agents.
- Build end-to-end systems that use agents for tasks such as kernel optimization, code generation, performance analysis.
Other
- 2+ years of software engineering experience.
- Strong communication skills, excellent problem-solving skills, and the ability to learn and work effectively in a fast-paced, collaborative environment.
- Applications for this job will be accepted at least until December 6, 2025.
- NVIDIA is committed to fostering a diverse work environment and proud to be an equal opportunity employer.
