Your future company
Devire IT Outsourcing is a form of cooperation dedicated to IT professionals, based on the principles of own business - B2B, implementing projects for clients running innovative and modern projects.
Our client is a technology innovator specializing in advanced computer vision solutions, including real-time image processing.
Their projects involve working with live camera feeds to deliver high-performance applications, such as anti-theft systems capable of tracking specific individuals from the moment they enter a store.
The solutions also integrate video streams from multiple cameras to create a unified view, enabling precise object tracking.
Additionally, the company develops AI-driven tools to analyze shelf inventory by recognizing and counting products directly from visual data, supporting retail efficiency and loss prevention.
Due to the dynamic growth of the organization, we are looking for candidates for the position of ML Engineer.
Requirements
Strong mathematical background (graph theory, optimization)
Experience with large-scale tracking problems
Proficiency in Python and PyTorch
Experience with TensorFlow
Familiarity with TensorRT (for model optimization), TensorFlow
Knowledge of GitLab, GitHub and Kubernetes (K8s)
English B2
Responsibilities
Design and optimize graph-based algorithms for multi-camera people tracking using frameworks like g2o.
Build robust multi-object, multi-camera tracking systems (MOMCT) using ReID and embedding-based association.
Train and implement state-of-the-art models (e.g., ViT-CLIP) using PyTorch.
Optimize trained models using TensorRT for efficient deployment.
The offer