Computer Vision Engineer, Healthcare AI

Computer Vision Engineer, Healthcare AI

Company: Ophthalytics

Employment Type: Full-Time, 40 hours per week

Start Date: ASAP

Compensation: Competitive salary based on experience, paid in U.S. dollars (USD)

Preferred Core Framework: TensorFlow 2.x / Keras

Help Us Build AI That Can Prevent Blindness Worldwide

Ophthalytics is a fast-growing U.S. healthcare AI company expanding globally. Our mission is to end preventable blindness by enabling earlier detection of vision-threatening diseases through advanced medical imaging and artificial intelligence.

Millions of people, especially in rural and underserved communities, do not have timely access to eye-care specialists. We are building AI technologies that can help close this gap, expand access to screening, and protect the vision of millions of people worldwide.

Ophthalytics is also developing Explainable AI, where algorithms do more than generate predictions. Our goal is to produce clinically meaningful and transparent outputs that support trust, understanding, and real-world adoption.

This Is Not a Regular Nine-to-Five Role

We are not looking for someone who only wants to complete assigned tickets and watch the clock. This is a mission-driven, fast-paced, high-ownership opportunity for an exceptional engineer who wants to build technology that matters.

You will join a select team of world-class AI and machine learning engineers and work directly with the Founder and top company leadership. You will help build advanced computer vision algorithms, extend production ML pipelines, influence technical decisions, and own important areas of product development from research through deployment.

This role offers a one-of-a-kind career experience. You will solve difficult technical problems, work on products with global impact, support underserved communities, and contribute to innovation in healthcare AI and other emerging technology areas.

Role Summary

We are hiring a Computer Vision Engineer with strong deep learning expertise and a proven ability to build production-ready machine learning systems on AWS.

You should have expert-level familiarity with modern computer vision frameworks and architectures, including TensorFlow, PyTorch, YOLO, EfficientNet, ConvNeXt, and Vision Transformers. You must be comfortable working across the complete ML lifecycle, including data preparation, model training, evaluation, optimization, deployment, and production monitoring.

This role has significant visibility and growth potential. Your work will directly influence product direction, technical priorities, and execution.

Key ResponsibilitiesComputer Vision Model Development

Build, train, evaluate, and improve models for:

• Image classification and multi-label classification
• Object detection
• Image segmentation
• Image quality assessment and automated quality gating
• Medical image analysis
• Explainable AI outputs

Implement and fine-tune modern architectures, including:

• YOLO and other state-of-the-art detection frameworks
• EfficientNet and ConvNeXt
• Vision Transformers, including ViT and Swin
• Emerging computer vision and multimodal architectures

Model Evaluation and Experimentation

Own model evaluation, benchmarking, and reporting using:

• F1 score and AUROC
• Sensitivity and specificity
• Calibration and threshold optimization
• Confusion matrix analysis
• False-positive and false-negative analysis
• Error analysis and subgroup performance evaluation
• Experiment tracking and model comparison

Data Management and Engineering

Design and maintain reliable ML data pipelines, including:

• Data ingestion, cleaning, normalization, and deduplication
• Dataset versioning, lineage, and metadata management
• Train, validation, and test split strategies
• Image labeling and annotation workflow support
• Dataset and label quality checks
• Reproducible data-processing and training pipelines
• Clear technical documentation for large image datasets

AWS Training and Deployment

Build scalable model-training and deployment workflows using:

• Amazon S3
• Amazon EC2
• AWS IAM
• Amazon CloudWatch
• GPU-based cloud infrastructure
• Production monitoring and reliability workflows

Model Productionization

Convert research models into reliable production systems using technologies such as:

• TensorFlow Serving
• TensorRT and TF-TRT
• TensorFlow Lite
• TorchScript
• ONNX

Improve:

• Inference latency and throughput
• GPU and compute utilization
• Scalability and availability
• Reliability testing
• Model versioning and release management
• Production performance monitoring

Software Engineering

Maintain strong engineering standards through:

• Clean, modular, and maintainable code
• Git and pull request workflows
• Code reviews
• Unit and integration testing
• Reproducible development environments
• Technical documentation
• Secure and reliable software architecture

Required Qualifications

• 3 or more years of hands-on experience in computer vision or deep learning, or an exceptional equivalent portfolio
• Strong experience with TensorFlow 2.x and Keras
• Working knowledge of PyTorch and other modern ML frameworks
• Strong Python programming skills
• Experience with classification, object detection, segmentation, and image-processing pipelines
• Experience with YOLO or similar object detection frameworks
• Strong understanding of loss functions, augmentation, regularization, optimization, and class imbalance
• Experience with calibration, thresholding, sensitivity, specificity, F1, AUROC, and error analysis
• Experience managing large image datasets and building reproducible ML pipelines
• Strong data management and data engineering capabilities
• Strong software engineering practices, including testing, documentation, code reviews, maintainable architecture, and Git
• Hands-on experience running ML workloads on AWS
• Working knowledge of AWS S3, EC2, IAM, and CloudWatch
• Ability to independently solve complex technical problems
• Strong communication, collaboration, accountability, and attention to detail

Preferred Qualifications

• AWS SageMaker
• Amazon ECR
• Amazon ECS or EKS
• AWS Step Functions
• Terraform or other Infrastructure as Code tools
• MLOps and automated ML pipelines
• Experiment tracking and model registries
• Model drift and performance monitoring
• Grad-CAM, saliency mapping, and attention visualization
• Explainable AI techniques
• Medical imaging or healthcare AI
• DICOM or DICOMWeb
• Clinical workflow experience
• Production deployment of GPU-based models
• Experience in regulated healthcare or life sciences environments

Who Will Succeed Here?

You will be a strong fit if you:

• Are deeply passionate about computer vision, AI, and healthcare
• Want your work to create measurable global impact
• Take full ownership of problems and outcomes
• Move quickly without compromising quality
• Thrive in a fast-growing startup environment
• Enjoy solving difficult and previously unsolved technical problems
• Can work closely with founders, technical leaders, and cross-functional teams
• Understand that successful AI requires excellent data, engineering, deployment, monitoring, and user trust, not only model accuracy
• Want to build real products, not just experiments

Why Join Ophthalytics?

• Global mission: Help prevent blindness through earlier detection of vision-threatening diseases.
• Meaningful ownership: Influence product development, architecture, experimentation, deployment, and continuous improvement.
• Direct leadership access: Work closely with the Founder and top company leadership.
• Exceptional team: Collaborate with world-class AI and ML engineers.
• Real-world impact: Build technology with the potential to reach millions, including underserved communities.
• Career growth: Grow into senior engineering, technical leadership, product ownership, or AI leadership roles.
• USD compensation: Receive a competitive salary paid in U.S. dollars.

Recognition and Momentum

Ophthalytics has been:

• Named one of Georgia’s Top 10 Most Innovative Companies in 2025
• Connected with world-class research ecosystems, including Georgia Tech
• Supported through the Microsoft startup ecosystem
• Supported as an AWS portfolio startup
• Recognized with an AWS Health Equity Award for real-world impact in rural communities
• Supported by NVIDIA
• Rapidly expanding across new customers, regions, partnerships, and product capabilities