Regular/Senior CNN ML Engineer

What you will do

We are looking for an ML Engineer focused on building fast, reliable inference systems and integrating machine learning into real-time graphics pipelines.

The project is positioned at the intersection of deep learning and computer graphics. The main focus is on Convolutional Neural Networks and high-performance inference integrated into rendering workflows. The work is closely connected with GPU technologies and real-time rendering systems.

You will collaborate with experienced engineers on advanced solutions for real-time image processing, rendering enhancement, and ML-driven graphics pipelines.

Design and build fast, reliable ML inference systems for real-time applications

Develop, train, and optimize Convolutional Neural Network (CNN) models

Integrate ML models into real-time graphics and rendering pipelines

Implement custom training loops and optimize training workflows

Apply distributed training and mixed precision techniques

Optimize model performance for GPU execution

Collaborate with graphics and low-level engineers on ML integration

Requirements

MUST

Strong deep learning background with solid expertise in Convolutional Neural Networks (CNNs)

• Advanced proficiency in Python and PyTorch
• Experience building and optimizing inference pipelines
• Ability to implement custom training loops
• Experience with distributed training and mixed precision
• Solid understanding of computer graphics fundamentals:
• Rendering pipeline
• Shaders
• Sampling
• Anti-aliasing
• Tone mapping
• Perceptual quality metrics
NICE TO HAVE

Experience with Vulkan or DirectX

Shader programming experience (HLSL / GLSL)

Experience integrating ML models into graphics pipelines

Rendering engine or game engine knowledge