Sr. Robotics Engineer - Perception and AI Fresh Consulting - 3.9 Bellevue, WA Job Details Full-time $125,000 - $150,000 a year 1 hour ago Qualifications AI models Robot integration Performance tuning Customer communication Software engineering Sensor calibration Azure Data transformation pipeline development Computer Science Data augmentation Machine vision systems Engineering development testing Requirements design Bachelor's degree in electrical engineering PyTorch Bachelor of Science Robotics engineering IoT systems Quantization Git Public Cloud AI platforms (beyond public GPTs) 3 years System design for system development AWS C++ Docker Professional development support Model deployment Master of Science Full Job Description We are Fresh. Together let's build the future. With partners and clients, we practice strategy, design, development, and engineering to harness the power of technology and create what's next. Our cross-disciplinary approach blends human ingenuity and technology, empowering us to meet uncertainty with confidence. What got us to today won't get us to tomorrow, so we test our assumptions and always strive for growth. From the products we build to the partners we collaborate with, we believe people make the difference. And we do so as a workforce representative of the communities we serve, understanding that a diverse workforce strengthens our organization. We value diversity and support a positive and welcoming environment where all of our employees can thrive. Why join Fresh? The Fresh Consulting hardware and robotics team is working with customers to build and deploy fleets of robots and intelligent distributed devices into the field. You'll own perception pipelines that bridge what our systems see and how they act, from sensor data collection through edge deployment. Our programs involve deploying in the real world at scale. We are looking for a Senior Robotics Engineer with a focus on perception and applied AI. While the methodology for building these systems has evolved rapidly, we have adapted alongside it. In addition to classical CV and purpose-trained models, we deploy vision foundation models and learned policies on active programs whenever they are the right fit for the challenge. Regardless of the approach, our standard remains consistent: the solutions we deliver must operate reliably once in the customer's hands. If you have a strong background in software engineering, want to work across various robot platforms and sensor stacks rather than a single one, and want to help shape and lead how intelligent robotic systems are built and deployed, we'd love to chat. Daily Responsibilities Design and own CV/ML pipeline architecture end-to-end: data collection, dataset curation, model training, event logic development, pipeline performance evaluation, and edge deployment. Select the optimal approach for each problem, utilizing classical CV and purpose-trained detectors where appropriate, or foundation models and learned policies (including VLA models) when task complexity makes scripted methods impractical. Develop, deploy and maintain the data engine pipeline, including auto-labeling, synthetic data generation from simulation, and field data observability loops to support continuous model performance improvements, using differentiated tactics and resource allocations to handle pre- and post-deployment requirements. Fuse data from RGB-D cameras, LiDAR, and IMUs, with deep awareness of each modality's tradeoffs, and familiarity of classical and learning-based multi-modal approaches, to fulfill specific technical goals. Maintain IoT and fleet data pipelines supporting OTA model updates and telemetry across deployed units. Define, analyze, and review perception requirements, verifying system performance through test automation and closed-loop evaluation. Consult on the suitability of newer methods, considering latency budgets, determinism, cost, and safety certification when favoring conventional approaches. Mentor engineers on the Fresh Robotics team, reviewing designs and code while helping them grow specifically in perception and ML disciplines. Collaborate with hardware, cloud, and software engineering disciplines to ensure perception stack compatibility and identify technical risks early for larger scope technical alignment. Participate occasionally in client scoping calls or customer site visits to assess feasibility and effort levels as a technical advisor. Skills and Qualifications

B.S./M.S.

in Computer Science, Robotics, Electrical Engineering, or a related field (or equivalent hands-on experience). 3+ years of professional experience in robotics or CV/ML Expert-level proficiency in Python and C++, with a track record of driving complex projects to completion. Deep experience with PyTorch across the full model lifecycle, including dataset work, training, and edge deployment. Hands-on experience with OpenCV and classical computer vision techniques. Deep experience with sensor fusion across camera, LiDAR, and IMU, including calibration and debugging on real hardware. Excellent understanding of ROS 2, including integration of perception outputs with navigation and planning stacks. Proficient with Docker, Linux command line, shell scripting, and Git workflows. Excellent oral and written communication skills, with the ability to explain technical tradeoffs to customer stakeholders. Preferred Qualifications Hands-on experience with vision foundation models, including open-vocabulary segmentation and detection, and judgment on model suitability. Experience with learned policies, including imitation learning, diffusion policies, or fine-tuning open VLA models. Experience optimizing large DNN models for edge compute using quantization, distillation, TensorRT, ONNX, and CUDA. Experience with synthetic data and sim-to-real tools such as Mujoco, Isaac Sim, Isaac Lab, or Gazebo. Experience building data engines involving auto-labeling, active learning, and closed-loop evaluation. Experience with mmWave Radar, SLAM, path planning, 3D geometry, and controls & dynamics. Familiarity with Kalman Filters and advanced state estimation techniques. Experience with IoT data pipelines, MQTT, and real-time or deterministic systems programming. Experience deploying algorithms for agent path-planning and verification of autonomous behaviors. Familiarity with mainstream cloud platforms such as AWS, GCP or Azure Knowledge of edge NPUs for power-constraint yet performance demanding sensor data processing tasks Experience leading or mentoring engineering teams in a professional or consulting context. For a Washington-State-based role, the base salary hiring range for this position is $125,000 to $150,000. Compensation offered will be determined by factors such as location, level, job-related knowledge, skills, and experience.

Equal employment opportunity:

All qualified persons will be considered for employment without regard to race, color, religion, sex, national origin, age, marital status, familial status, gender identity, sexual orientation, disability for which a reasonable accommodation can be made or any other status protected by law. Assistance will be gladly provided upon request for any applicant with sensory or non-sensory disabilities. Fresh Consulting is an E-Verify participating company AgVY8JGmXh