C++ ENGINEER WITH COMPUTER VISION
Svitla Systems · Remote - Poland · 2026-08-07
Job description
Svitla Systems Inc. is looking for a Senior C++ Engineer with Computer Vision for a full-time position (40 hours per week) in Poland. Our client is an American mobile computing company.
You will bridge the gap between high-level machine learning research and high-performance production environments. Also, you will be responsible for building robust SDKs, automating deployment pipelines, and ensuring the models run efficiently across a diverse hardware landscape, from edge SoCs to powerful cloud GPUs.
Requirements
- Bachelor's or master's degree in Computer Science, Electronic Engineering, or related technical field.
- Deep understanding of C++14/17/20, including STL, memory management, and multi-threading.
- Expertise in writing clean, maintainable Python for automation and data processing.
- Hands-on experience with at least one of the following frameworks:
- SNPE/QNN (Qualcomm);
- OpenVino (Intel);
- TensorRT (NVIDIA);
- TensorFlow Lite.
- Familiarity with Docker for creating consistent development and deployment environments.
Nice to have
- Understanding of deep learning fundamentals (CNNs, Transformers, Object Detection).
- Knowledge of model conversion and quantization (i.e., PTQ, QAT).
Responsibilities
- SDK Development (C++): Design, develop, and maintain high-performance software development kits (SDKs) to expose computer vision capabilities to end users and internal products.
- Model Deployment and Integration: Port, convert, and deploy machine learning models across various hardware targets, including Qualcomm SoCs, Intel CPUs, and NVIDIA GPUs.
- Performance Optimization: Use hardware-specific toolkits to optimize model throughput without sacrificing accuracy.
- Evaluation & Benchmarking: Conduct rigorous testing and evaluation of models on target hardware to ensure performance metrics meet expectations.
- Automation: Build and maintain automation scripts and CI/CD pipelines in Python to streamline the model testing and deployment lifecycle.