Deep Learning Architect, Llm Inference - New College Grad 2026
NVIDIA
Base: $124,000 - $195,500 (level 2) or $152,000 - ...
**
Deep learning inference serving expertise
Pytorch programming and compiler optimizations
Cpu and gpu microarchitecture knowledge
**
NVIDIA is seeking a Deep Learning Architect focused on LLM Inference for new college graduates in 2026. The role involves optimizing inference server performance for large language models, collaborating with various teams, and contributing to deep learning projects while fostering a culture of innovation and excellence.
**
Job Summary
The role focuses on workload characterization of the latest Large Language Models and inference servers like vLLM, SGLang, and TRT-LLM.
Candidates will collaborate with engineers from AI startups to establish standard benchmarking methodologies and contribute to deep learning software projects.
NVIDIA offers a competitive base salary range of $124,000 to $241,500 USD along with equity and benefits for this position.
Matching Summary
Match Score: 75
**
NVIDIA is seeking a Deep Learning Architect focused on LLM Inference for new college graduates in 2026. The role involves optimizing inference server performance for large language models, collaborating with various teams, and contributing to deep learning projects while fostering a culture of innovation and excellence.
**
Salary
Base: $124,000 - $195,500 (Level 2) or $152,000 - $241,500 (Level 3); Bonus/Equity: Eligible for equity; Benefits: Comprehensive benefits package included
Skills & Requirements
Must-have
Deep learning inference serving expertise
PyTorch programming and compiler optimizations
CPU and GPU microarchitecture knowledge
Client server LLM application development
Profiling and performance bottleneck identification
Nice-to-have
Experience with agentic AI tools usage
Database and visualization tool proficiency
Novel use cases for workplace AI agents
Strong written and verbal communication skills
Proactive independent approach to problem solving
Key Requirements
Master's or PhD in Computer Science or related field