Robotics Perception Architect

About the job About The Role You will join a long-term initiative in collaboration with a globally recognized industry leader. The project focuses on advancing physical AI and multi-fleet management systems to significantly enhance the performance and efficiency of outdoor vehicles. This is a unique opportunity to work at the intersection of robotics, AI, and real-world hardware, where solutions go beyond simulation into practical application. Responsibilities Define and own the end-to-end architecture of robotics perception systems, covering sensor selection, data pipelines, and real-time processing workflows Design scalable, modular perception solutions integrating computer vision, sensor fusion, AI/ML, Physical AI, and localization across edge and cloud environments Collaborate with clients and cross-functional teams to translate business requirements into architecture blueprints, roadmaps, and technology decisions Establish AI/ML strategies for perception, including model selection, data lifecycle management, and deployment aligned with performance requirements Integrate perception components into broader robotics systems — navigation, manipulation, planning — ensuring clean interfaces and interoperability Drive performance optimization by balancing latency, accuracy, and resource constraints in real-world and safety-critical conditions Lead engineering teams through design reviews, mentoring, and architectural standards enforcement across the full delivery lifecycle Explore emerging technologies, validate new approaches, and build reusable accelerators that strengthen the robotics consulting practice Requirements 8+ years of software development experience in C/C++ and Python, including 4+ years in lead or architect roles Hands-on expertise with AI-based robotics perception, computer vision, and agentic systems in production environments Strong command of ML frameworks and robotics tooling: TensorFlow, PyTorch, OpenCV, ROS/ROS2, object detection, segmentation, tracking, and Point Cloud/RGB-D processing Solid experience with SLAM, localization, mapping, motion planning for mobile platforms, and trajectory planning for robotic arms using MoveIt Practical knowledge of simulation and synthetic data generation tools such as Isaac Sim, Blender, or Genesis Experience deploying AI and CV models to production, including MLOps, CI/CD pipelines, and monitoring for complex systems Familiarity with cloud platforms (AWS, Azure) and Unix/Linux environments Strong mathematical foundation with the ability to design and implement efficient algorithms and software architectures SoftServe is an equal opportunity employer. Qualified applicants will receive consideration regardless of race, color, ancestry, ethnicity, national origin, religion, sex, sexual orientation, gender identity or expression, age, citizenship, disability, health condition, marital or family status, veteran status, or any other characteristic protected by applicable law.