Computer Vision Engineer
We are looking for a Senior/Lead Computer Vision Engineer to join Ralabs and work on technically challenging projects involving edge computing and AI-powered solutions.
The role is focused on computer vision: object detection models, real-time video pipelines, and inference optimization. Our solutions run on edge devices, so you should understand how embedded systems work and how hardware limitations affect model and pipeline design. You will work closely with embedded engineers, who handle low-level hardware integration.
Beyond individual projects, you will contribute to our broader engineering capabilities in computer vision and edge AI.
📍 Location: Ukraine-based
🕒 Engagement: Part-time (approximately 10–20 hours per week during the first 1–3 months), with the potential to transition to full-time.
Requirements
Experience
- 6+ years of commercial software engineering experience, including 4+ years in computer vision / ML.
- Proven experience optimizing CV/ML models for inference performance and resource usage.
- Experience deploying or running CV models on edge devices (e.g., NVIDIA Jetson, Raspberry Pi, or similar).
- Ability to independently investigate technical challenges and deliver working solutions.
Technical Expertise
Computer Vision & ML
- Strong proficiency in Python; working knowledge of C/C++ is a plus.
- Hands-on experience with YOLO-NAS, YOLOX, and Darknet YOLO models in ONNX format.
- Strong background in computer vision, image processing, and real-time video processing.
- Experience with model optimization for edge deployment, including quantization, pruning, and conversion (ONNX, TensorRT, or similar).
- Experience with OpenCV and real-time video pipelines.
Embedded Understanding
- Understanding of how embedded systems work and how their constraints (CPU/GPU, memory, power and thermal) affect inference.
- Understanding of the end-to-end data flow on a device, from camera capture and pre-processing to inference and output.
- Working knowledge of Linux.
- Ability to communicate effectively with embedded engineers about hardware requirements and limitations.
General
- Familiarity with software security and secure deployment practices.
- Strong understanding of performance optimization, testing, and troubleshooting.
Leadership & Collaboration
- Ability to take technical ownership of assigned tasks and solutions.
- Strong analytical and problem-solving skills.
- Ability to evaluate technical approaches (models, architectures, trade-offs) and make practical engineering decisions.
- Comfortable working with incomplete requirements and exploring alternative solutions.
- Ability to collaborate with cross-functional engineering teams, including embedded engineers, and technical stakeholders.
Communication
- English proficiency: Upper-Intermediate.
- Ability to clearly communicate technical decisions, risks, and dependencies.
- Strong technical documentation skills.
Responsibilities
Computer Vision & Edge AI
- Select, train, fine-tune, and evaluate CV models (object detection, tracking, etc.) for edge use cases.
- Convert and optimize models for target edge hardware.
- Build and improve real-time computer vision and video processing pipelines.
- Balance model accuracy against inference speed and resource usage.
- Evaluate technical approaches for improving system performance.
Edge Deployment
- Choose model architectures and optimization strategies with the target hardware’s limitations in mind.
- Test and profile models on edge devices together with the embedded team.
- Work closely with embedded engineers on integrating CV models into the device software.
Engineering Excellence
- Write clean, maintainable, and well-tested code.
- Establish reliable model evaluation, testing, and deployment practices.
- Identify performance bottlenecks and propose improvements.
- Ensure appropriate security and reliability standards.
- Maintain technical documentation.
Collaboration & Delivery Support
- Work closely with engineering teams, technical leads, and project stakeholders.
- Participate in technical discovery and solution design.
- Support technical estimations and feasibility assessments.
- Communicate technical challenges, risks, and dependencies.
- Take ownership of delivering working, production-ready solutions.
Pre-Sales & Solution Design (Optional Involvement)
- Support technical discovery for new computer vision and edge AI opportunities.
- Evaluate the feasibility of CV solutions on target hardware.
- Contribute to solution design and effort estimation.
- Participate in technical discussions with prospective clients.
Technology Office Contribution
- Help develop Ralabs’ expertise in computer vision and edge AI.
- Research and evaluate relevant models, frameworks, and hardware platforms.
- Share technical knowledge and engineering best practices.
- Contribute to reusable solutions and internal engineering standards.
Would be a plus
- Hands-on experience with embedded Linux and ARM-based platforms.
- Experience with CUDA, TensorRT, DeepStream, or other NVIDIA technologies.
- Experience with GStreamer.
- Understanding of hardware interfaces (MIPI CSI, USB cameras, UART, GPIO, SPI, I2C).
- Experience with cross-compilation and deployment to target devices.
- Experience with model training pipelines and dataset preparation or annotation.
- Experience with multi-camera systems or object tracking.