Multimodal Inference Engineer - Real-Time AI Systems

OpenAI; California (MO); On-site; USD 180,000 - 240,000; Full time

Job summary

OpenAI is seeking a software engineer to scale multimodal model inference. You will build reliable, high-performance infrastructure for serving real-time audio, image, and other modalities in production. You will work cross-functionally with researchers, product teams, and infra to deploy state-of-the-art capabilities.

You should thrive in fast-moving spaces, own end-to-end problems, and have experience with GPU-based ML workloads and inference tooling like vLLM or TensorRT-LLM.

Qualifications

• Experience building and scaling inference systems for LLMs or multimodal models.
• Experience with GPU-based ML workloads and performance of large models.
• Familiarity with inference tooling such as vLLM or TensorRT-LLM.

Responsibilities

• Design and implement inference infrastructure for large-scale multimodal models.
• Optimize systems for high-throughput, low-latency delivery of image and audio inputs/outputs.
• Enable experimental research workflows to production services.
• Collaborate with researchers, infra teams, and product engineers to deploy capabilities.
• Improve system performance including GPU utilization and hardware abstraction.

Skills

Inference systems

Multimodal ML

LLMs

GPU workloads

System optimization

Tools

vLLM

TensorRT-LLM

Model parallel

Job description

OpenAI is seeking a software engineer to scale multimodal model inference. You will build reliable, high-performance infrastructure for serving real-time audio, image, and other modalities in production. You will work cross-functionally with researchers, product teams, and infra to deploy state-of-the-art capabilities.

You should thrive in fast-moving spaces, own end-to-end problems, and have experience with GPU-based ML workloads and inference tooling like vLLM or TensorRT-LLM.