Computer Vision Engineer

Job Title: Computer Vision Engineer Company: Eternal Job Category: Machine Learning Roles & Responsibilities: About the Role ------------------ * Maintain and improve our perception pipeline, including detecting ripe tomatoes among dense foliage, estimating cut points on stems with sub\-centimeter accuracy, and adapting to new crop types as we expand. * Build and maintain data and model lifecycle infrastructure, from data collection and annotation to model monitoring, drift detection, and retraining across our growing robot fleet. * Create real\-time perception pipelines that process 2D/3D sensor data for robotic decision\-making with sub\-centimeter accuracy. * Optimize vision algorithms for edge deployment on robotic platforms, balancing accuracy with computational efficiency. * Build tools for debugging, analysis, and performance evaluation of your algorithms. Root\-cause failures when things break in production. * Collaborate cross\-functionally with robotics engineers, AI/ML researchers, and crop scientists to deliver end\-to\-end perception solutions. * You own your projects from prototype to production. We expect production\-quality software engineering, not just notebook experiments. We don't have separate research and engineering teams Rapid Innovation: Our software\-first approach means you'll see your models deployed to real robots in hours/days, not months. We've proven we can develop and deploy new perception capabilities over\-the\-air as crops evolve. Flexible Work Culture: Distributed team with offices in Cologne and Bengaluru, following a "follow\-the\-sun" support model for our 24/7 operations. eternal.ag* is building fully automated food production sites that can sustainably produce fresh food year\-round. Backed by world\-class investors and partnering with leading agricultural companies, we're turning the vision of fully\-autonomous greenhouses into reality. Work Location: Karnataka, India Availability: Full time

Help us bring our robots to life in greenhouses around the world. Whether you're a recent graduate or an experienced engineer, you'll build perception systems that enable robots to identify ripe produce, detect plant diseases, and navigate complex greenhouse environments autonomously. #### Core requirements (All levels) * Bachelor's or Master's degree in Computer Science, Electrical Engineering, Applied Mathematics, Artificial Intelligence, Machine Learning, Robotics, or a related field (or graduating by Summer 2026\) * Strong programming skills in C\+\+ and/or Python * Solid understanding of deep learning fundamentals (training, evaluation, common failure modes) and how to apply them to real\-world problems * Understanding of 3D geometry, camera models, and coordinate transforms * Working knowledge of classical CV (OpenCV, image processing, filtering) * Experience with at least one deep learning framework (PyTorch preferred) * Familiarity with Linux environments and version control systems * A bias toward letting data and learning solve problems, while being pragmatic about when classical approaches are the right tool #### Experience levels New Graduate / Entry Level (0\-2 years) * Recent graduate or final year student with strong academic performance * Hands\-on computer vision and/or deep learning experience through internships, research projects, or competitions (Kaggle, university labs, personal projects) * Demonstrated programming skills through coursework or personal projects * Understands model training and evaluation basics, including common failure modes * Has built a training pipeline end\-to\-end: data collection/curation, training, evaluation, deployment, and iteration based on real\-world feedback * Experience taking at least one vision or ML system from prototype to production * Proficiency with modern architectures (YOLO, Mask R\-CNN, Vision Transformers) * Can make sound tradeoffs between classical and learning\-based components in a perception pipeline * Practical experience with model optimization for edge deployment (quantization, distillation, TensorRT/ONNX export) * Proven track record of deploying production perception systems and making sound architectural tradeoffs between classical and learning\-based approaches * Experience designing data strategies: what to collect, how to annotate efficiently, when to retrain vs. fine\-tune * MLOps experience: automated retraining pipelines, model versioning, drift detection, or A/B evaluation across a deployed fleet * Knowledge of model optimization for embedded systems (quantization, pruning, distillation) * Ability to mentor junior engineers and make technical decisions that compound over time Launch Your Career: For new graduates, this is a unique opportunity to join a proven team and learn from engineers who've already built and deployed commercial vision systems. You'll get hands\-on experience with cutting\-edge technology while making a real\-world impact from day one. Unique Challenges: Tackle perception problems that combine the complexity of outdoor vision (varying lighting, weather) with the precision requirements of industrial automation. Growth Opportunity: Join as we scale from proof\-of\-concept to global deployment. Be part of the core team shaping the future of agricultural perception. Clear career progression from graduate to senior engineer and beyond. We're committed to building a diverse and inclusive team. We encourage applications from candidates of all backgrounds and experience levels who are excited about our mission and show potential to grow with us. Recent graduates \- don't let experience requirements hold you back; we value passion, potential, and fresh perspectives. Key Skills: Machine Learning, Python, PyTorch, Model Training, Computer Vision Salary: Not disclosed