Senior Computer Vision Engineer

Aptly Automation is a forward-thinking technology company dedicated to addressing complex real-world challenges through advanced computer vision and data science. Our multidisciplinary team collaborates to develop scalable, production-ready systems that power intelligent products and services. We foster a culture of curiosity, rigorous engineering, and continuous improvement in both research and development. Operating with a fully remote team, we emphasize flexibility, inclusivity, and access to diverse talent. Joining Aptly Automation means working with cutting-edge tools, contributing to impactful solutions, and growing within a supportive environment.

Key Responsibilities
Design, implement, and optimize computer vision models for tasks such as image and video analysis, classification, object detection, tracking, and recognition. Develop and refine image processing and video-AI pipelines, ensuring reproducibility through configuration management, hyperparameter tuning, and experiment tracking. Evaluate model performance rigorously and enhance accuracy via iterative data refinement and retraining cycles. Export and optimize models for production deployment, balancing accuracy, latency, throughput, and infrastructure costs. Build and maintain production-grade computer vision services and APIs deployed on cloud and edge platforms. Own solutions end to end, from prototype to production, collaborating closely with cross-functional teams to define requirements and integrate vision components into larger systems. Conduct research and prototype innovative approaches to solve challenging problems. Manage and curate image and video datasets, including annotation tasks. Review and improve existing codebases, contribute to architectural decisions, and uphold engineering best practices. Mentor team members, conduct code reviews, and maintain clear technical documentation. Contribute to the company’s long-term technology strategy.

Required Qualifications
Bachelor’s degree in Artificial Intelligence, Computer Science, or related fields such as Electrical Engineering or Data Science; Master’s degree is a plus. A minimum of 5 years of professional experience in computer vision and machine learning, including designing, training, and deploying models in production environments. Proficient in Python with strong software engineering fundamentals; C++ experience is advantageous. Hands-on expertise with PyTorch and associated tools like torchvision and PyTorch Lightning; familiarity with TensorFlow/Keras is beneficial. Practical experience in image classification (CNNs, transfer learning), object detection frameworks such as YOLOv8 (Ultralytics), Faster R-CNN, Mask R-CNN, RT-DETR, and U-Net. Knowledge of Detectron2, MMDetection, or Segment Anything (SAM) is a plus. Experience with video AI pipelines and multi-object tracking methods like ByteTrack and DeepSORT. Proficiency in computer vision libraries including OpenCV, scikit-image, and Pillow, along with classical image processing techniques such as feature extraction (SIFT, ORB), filtering, and edge detection. Strong foundation in machine learning and deep learning concepts, including CNNs, Vision Transformers, transfer learning, loss functions, optimization, and regularization. Skilled in dataset handling and augmentation using tools like Albumentations, torchvision transforms, and Kornia, alongside rigorous model evaluation metrics such as precision, recall, mAP, and IoU. Experience building reproducible training pipelines and experiment tracking with tools like Hydra, Optuna, MLflow, or Weights & Biases. Competence in exporting models to ONNX and serving them via APIs (e.g., FastAPI) with inference optimization. Familiarity with data science libraries such as NumPy, Pandas, scikit-learn, Matplotlib/Seaborn, and Jupyter notebooks. Excellent English communication skills, both written and spoken, and the ability to work independently in a remote setting. Strong analytical and problem-solving skills with ownership of projects from start to finish.

Preferred Qualifications and Benefits
Experience with model deployment and optimization tools including ONNX Runtime, TensorRT, OpenVINO, TensorFlow Lite, NVIDIA Triton, model quantization, pruning, performance profiling, and CUDA fundamentals. Familiarity with serving frameworks such as FastAPI (async), Flask, Pydantic, and REST API design. Knowledge of MLOps tools and workflows including Docker, Kubernetes, Hydra, Optuna, MLflow, Weights & Biases, TensorBoard, DVC, and CI/CD pipelines with GitHub Actions. Cloud experience with AWS SageMaker, GCP Vertex AI, or Azure ML, including GPU-based and distributed training. Proficiency with annotation tools like CVAT, Roboflow, Label Studio, or LabelImg. Exposure to advanced vision tasks such as instance segmentation, anomaly detection, OCR, document understanding, and generative models including diffusion models and vision-language models. Experience collaborating in version-controlled, multidisciplinary teams and prior leadership or mentoring roles.

Tech Stack Highlights
Python, C++, SQL; PyTorch, PyTorch Lightning, TensorFlow/Keras; YOLOv8, Ultralytics, Detectron2, MMDetection, SAM; OpenCV, scikit-image, Pillow, Kornia, Albumentations; NumPy, Pandas, scikit-learn, Matplotlib, Seaborn; ONNX, TensorRT, OpenVINO, Triton, CUDA; FastAPI, Flask, REST APIs; Docker, Kubernetes, Git, Hydra, MLflow, Weights & Biases; CVAT, Roboflow; AWS, GCP, Azure cloud environments.

What We Offer
A fully remote, full-time position with opportunities to work on innovative projects using state-of-the-art tools. A supportive culture that values curiosity, rigorous engineering, and continuous learning. Opportunities to lead technical strategy and mentor a growing team within a dynamic and inclusive environment.