About Rendair

Rendair is a visualization platform built for architects, interior designers and real estate professionals. Our tools combine advanced AI models with intuitive workflows, so users generate high quality images, animations and design variations in seconds. More than a million people in 180 countries use it, and the team behind it is small, remote and growing. We are continuously improving our AI infrastructure to deliver faster, more scalable and production ready image generation.

The role

We are looking for an experienced AI Infrastructure Engineer to help us design, optimize and scale our GPU based image generation pipeline. This role focuses on production level ML deployment and performance. You will build a stable, efficient inference architecture based on Diffusers and PyTorch, replacing workflow based systems with a more scalable infrastructure.

What you will do

Design and optimize a production ready Diffusers inference pipeline
Deploy and manage Dockerised GPU environments (CUDA, PyTorch)
Optimize model loading times and VRAM usage
Build scalable async job queue systems for image generation
Improve inference speed, reliability and cost efficiency
Support multi model workflows
Contribute to long term AI infrastructure decisions

What we look for

Strong experience with PyTorch
Hands on experience with HuggingFace Diffusers
Deep understanding of CUDA and GPU memory management
Experience deploying ML models inside Docker containers
Experience building scalable backend inference services
Strong debugging and optimization skills

Bonus: experience with SDXL, Flux or LoRA implementations, with GPU cloud providers such as RunPod or AWS, and with optimizing inference pipelines for production.

How we work together

Full time freelance collaboration, remote, high ownership and autonomy, with a long term collaboration in mind.