Note: The job is a remote job and is open to candidates in USA. Stability AI is seeking a passionate Generative AI Inference Engineer to join their Inference team, focusing on creative applications of generative AI models. The role involves leading the design and development of customer-facing multi-modal ML inference systems and optimizing inference techniques for generative models.

Responsibilities

• Lead efforts to drive the design, development of customer-facing multi modal ML inference systems
• Work with the Platform and Inference teams on building inference systems for the next generation of models, where you will work on areas such as optimization, model tuning and deployment
• Partner with leading cloud providers to deliver hosted Stability AI inference solutions
• Be a strategic thought partner for leaders across the organization on driving business impact through machine learning
• Be part of the team to bring new Stability models and pipelines into existence
• Prototype and productionize inference platform improvements and new features

Skills

• 7+ years working on productionizing machine learning systems, including inference pipeline development
• Expert level knowledge on writing and running python services at scale
• 5+ years working on python scientific stack, pyTorch and at least one high-performance inference framework (e.g. Triton and TensorRT)
• Deep understanding of Diffusion Architecture
• Experience profiling and optimizing deep neural networks on Nvidia GPUs, using profiling tools such as NVIDIA Nsight
• Experience with python-based image manipulation/encoding/decoding frameworks, such as OpenCV
• Experience deploying to cloud orchestration systems such as Kubernetes and cloud providers such as AWS, GCP, and Azure
• Experience with Docker
• Ability to rapidly prototype solutions and iterate on them with tight product deadlines
• Strong communication, collaboration, and documentation skills
• Experience with the open-source ML ecosystem (HuggingFace, W&B, etc.)

Company Overview