Machine Learning Engineer (Edge AI & Computer Vision)

Skills

What you'll do

• Take end-to-end ownership of Machine Learning models, moving them beyond training and ensuring reliable deployment in production on physical hardware.
• Design, build, and optimize real-time AI pipelines, integrating foundation models, sensor and application data, and scalable inference workflows.
• Optimize inference performance, memory usage, and execution speed for Edge AI platforms (e.g., NVIDIA Jetson, embedded platforms).
• Collaborate directly with clients and cross-functional engineering teams, building trust, proposing proactive solutions, and maintaining smooth technical communication.
• Debug, monitor, and maintain production models operating continuously under real-world hardware constraints.
• Strong experience bringing AI to production: It’s not just about training models—you know how to make them run reliably in real-world environments.

About the role

Want to work with cutting-edge technologies on long-term, strategic projects that combine Deep Learning, Computer Vision, Edge AI, and IoT ? 🚀🤖🚜 At Marvik , we are leading the next evolution of smart machinery—integrating vision capabilities, real-time image processing, and multi-sensor fusion into physical products operating in real environments. 🌟 What do we offer?: • Challenging, real-world projects: Work on stuff other people only read about—bringing AI out of the cloud and straight into physical, intelligent machinery. • State-of-the-art tech stack: Optimize and deploy cutting-edge Computer Vision and Deep Learning models on Edge devices. • Strategic growth: Join an expanding team with huge potential for long-term ownership, leadership, and professional development. • Great team culture: Excellent work environment full of highly motivated, collaborative, and talented professionals who elevate each other. • Global exposure & On-site experiences: Opportunity to travel internationally to work directly with clients. • Flexible work style: Opportunity to work remotely with global, high-impact clients. 🧑🏻‍💻 Responsibilities: • Take end-to-end ownership of Machine Learning models, moving them beyond training and ensuring reliable deployment in production on physical hardware. • Design, build, and optimize real-time AI pipelines, integrating foundation models, sensor and application data, and scalable inference workflows. • Optimize inference performance, memory usage, and execution speed for Edge AI platforms (e.g., NVIDIA Jetson, embedded platforms). • Collaborate directly with clients and cross-functional engineering teams, building trust, proposing proactive solutions, and maintaining smooth technical communication. • Debug, monitor, and maintain production models operating continuously under real-world hardware constraints. 🤝 If you have: • Strong experience bringing AI to production: It’s not just about training models—you know how to make them run reliably in real-world environments. • Solid background in AI & Deep Learning , including experience with LLMs, Computer Vision, multimodal models, model optimization, and efficient inference. • Edge AI & Embedded exposure: Experience optimizing models for constrained devices (TensorRT, ONNX, OpenCV, C++, or Python). • Strong Ownership & Soft Skills: Highly collaborative, proactive, independent, and clear in technical communication. A team player who builds trust with clients and peers. • Advanced English level: Excellent verbal and written communication skills to interact directly with international client teams. • Required tools: Python, C++, PyTorch/TensorFlow, OpenCV, Docker, Git. 🦾 It’s a major plus: • Hands-on experience with NVIDIA Jetson , ROS/ROS2, or embedded hardware platforms. • Experience in domains such as Robotics, IoT, Drones, Automotive, or Industrial Machinery . • Knowledge of sensor fusion (IMU, cameras, LiDAR) or OTA (Over-The-Air) updates and Cloud IoT architectures (AWS/Azure IoT). Originally posted on Himalayas