Nvidia Senior Software Engineer Agentic AI 

United States, Texas 

381584343

What you'll be doing:

• Implementing new features of our GenAI SDKs that enable LLM agents to expand to new, more demanding use cases and larger deployment configurations.

• Crafting proof-of-concept workflows rooted in first principles that apply modern data science techniques to GenAI use cases.

• Collaborating with other engineers to develop new optimizations for agentic applications across the entire data center, which focus on improving accuracy, reducing latency, and growing efficiency.

• Building integrations between the AIQ toolkit and other NVIDIA products and services, such as the NeMo Framework, NIMs, and NVIDIA Blueprints.

• Working with data scientists and ML/DL engineers to move from proof-of-concept analysis and modeling to production-ready pipelines and deployments.

What we need to see:

• BS in Computer Engineering, Computer Science, Data Science, or other closely related field (or equivalent experience).

• Proficient in Python, with at least 5+ years of experience building Python libraries or applications for enterprise customers.

• Experience with GenAI application development using LLM frameworks (such as Langchain, Llamaindex, or AutoGen), evaluation systems (such as RAGAs), and observability platforms (such as Arize Phoenix, W&B Weave, or LangSmith).

• Understanding of different agent architectures, RAG systems, and communication protocols (such as MCP or Google A2A).

• Deep desire to solve complex engineering challenges with efficiency as a priority.

• Ability to quickly learn and apply new technologies and libraries.

• Self-starter with a proactive attitude, capable of working independently and effectively within a distributed team.

• Excellent communication skills, essential for collaboration with multi-functional teams.

Ways to stand out from the crowd:

• MS, PhD or equivalent experience in Computer Engineering, Computer Science, Data Science, or other closely related field.

• Experience developing for GPU platforms and familiarity with NVIDIA technologies (e.g., CUDA, TensorRT, Triton, NeMo) and LLM serving frameworks (e.g., Dynamo, vLLM, SGLang).

• Proficient in distributed systems and communication frameworks (e.g., Ray, Dask, Spark, gRPC, Kafka, nats.io).

• Proven ability to prototype and productionize features, including deploying large-scale agentic applications with high concurrency.

• Track record of contributing to open-source Python projects.