Principal AI Platform Engineer
Summary
Hands-on Principal Engineer building an AI platform for deploying and operating LLMs in production at scale across multi-GPU, multi-node environments. Day to day: Python services, tuning vLLM model serving, optimizing GPU memory, latency and throughput, and automating infrastructure, deployment and reliability.
We’re looking for a Principal Software Engineer to help build a platform that brings AI models into production. The product enables engineering teams to deploy and operate LLMs at scale, handling real-world traffic across multi-GPU and multi-node environments.
You’ll work at the intersection of Python development and AI infrastructure, making inference services faster, more efficient, and reliable in production. This is a hands-on role with ownership across development, deployment, and operations.
What you’ll do:
Build Python services and tools for deploying and operating AI workloads.
Scale and optimize LLM inference across multiple GPUs and compute nodes.
Configure and tune model-serving frameworks such as vLLM.
Improve inference performance, GPU memory usage, latency, and throughput.
Automate infrastructure, deployment, and configuration workflows.
Strengthen monitoring, testing, and production reliability.
Troubleshoot complex issues across software and infrastructure.
What we’re looking for:
Strong Python development experience.
Hands-on experience running LLM inference services in production.
Experience with multi-GPU and multi-node model deployment.
Practical knowledge of vLLM or similar serving frameworks.
Understanding of GPU memory constraints and inference optimization techniques.
Experience with infrastructure automation and Infrastructure as Code.
Strong Linux troubleshooting and production operations skills.
Nice to have: Distributed systems, Kubernetes, cloud infrastructure, networking, or experience building AI platforms and developer tools.
