Candidate Profile — Candidate #1
Position: FieldAI — Robotics AI Engineer (Irvine, CA)
Candidate Location: Philadelphia, PA Metro
Experience: 6.7 years experience
Relevant Experience & Education Highlights
Excelled in developing ROS2-based autonomous systems, CUDA-optimized AI pipelines, and hands-on robotics deployments, aligning closely with FieldAI's needs for risk-aware Field Foundation Models and Dynamics Foundation Models.
1. Algorithm Development
Pioneered AI-driven models, planning algorithms, and trajectory optimization for autonomous robotics in unstructured environments.
- Designed and implemented ROS2 software stack for quadrupedal robotic prototype, focusing on SLAM, perception, and path planning algorithms.
- Developed multi-modal recommendation engine and sensor fusion pipelines using YOLOv8, MediaPipe, ResNet50, YOLOv5, Grounding DINO, and SAM, achieving 0.2% error rate and 80% reduction in manual intervention.
- Engineered deep learning control for ankle exoskeletons using TCN, autoencoders, VAEs, and GANs to predict gait phases and generate biologically inspired torque.
- Optimized GPT-4 inference latency by 55% via CUDA kernel fusion and TensorRT quantization for processing 50K+ daily requests.
2. Software Integration and Testing
Integrated next-gen autonomous modules with ROS, established real-world learning loops, and validated performance through field tests.
- Architected ROS2 middleware controlling IoT-enabled robotic fulfillment systems, including multi-node visual odometry for point cloud construction.
- Automated QA pipelines with YOLOv8 vision models and built CI/CD with GitHub Actions, achieving 98% test coverage and 40% cloud cost reduction.
- Migrated microservices to containerized environments with Kubernetes autoscaling, Terraform, Docker, and AWS ECS, ensuring 99.9% uptime.
- Implemented real-time dashboards in React for batch tracking, reducing operational overhead by 40%.
3. Hands-On Robotics Deployment
Collaborated on hardware-software integration, deployed systems in diverse conditions, and worked with legged and manipulator robots.
- Developed sensor fusion and integration for aerial drones with manipulators using RGB-D cameras and GPS for obstacle avoidance and object pick/drop.
- Designed modular pipe climber robot with mechanical fabrication and control autonomy for industrial applications.
- Contributed to control autonomy for robotic prototypes at research labs, including exoskeleton deployment and quadrupedal systems.
- Established MLOps foundations with spot instance orchestration and cost monitoring, saving significant cloud expenses.
Requirements & Candidate Alignment
| FieldAI Requirement | Candidate Qualification |
|---|---|
| Strong programming skills: Proficiency in Python and C++ for algorithm development and system integration. | Proficient in Python and C++, demonstrated through ROS2 middleware, CUDA C++ kernel fusion, TensorRT, and deep learning implementations. |
| Proficiency in ROS (Robot Operating System): Experience developing and integrating robotic software solutions. | Extensive ROS2 experience, including full software stack for quadrupedal robots, middleware for fulfillment systems, and SLAM/path planning modules. |
| Hands-On Robotics Experience: Practical experience working with autonomous robotic systems, including testing and deployment in the field. | 6.7 years hands-on with autonomous systems, spanning quadrupedal prototypes, exoskeletons, aerial drones, manipulators, and pipe climbers. |
| Understanding of Planning and Controls: Solid knowledge of path planning, control theory, and trajectory optimization. | Strong planning and controls expertise, via SLAM, visual odometry, path planning, TCN-based gait prediction, and torque control for exoskeletons. |
| Experience with GPU programming (CUDA C++, PyTorch): A significant plus. | GPU programming with CUDA C++, PyTorch, and TensorRT, optimizing inference latency by 55% and fusing kernels for high-throughput AI pipelines. |
| Knowledge of machine learning techniques applied to robotics: | Applied ML including YOLOv8, TCN, autoencoders, VAEs, GANs, for perception, anomaly detection, and control in robotic systems. |
| Experience working with humanoid or legged robots: | Worked with legged robots, developing ROS2 stacks for quadrupedal prototypes and ankle exoskeletons. |
| Advanced degree (Bachelors, Master's) in Robotics, Computer Science, Electrical Engineering, or a related field: | MS in Robotics from a top research university and BS in Mechanical Engineering from a reputable engineering college. |
If you would like to discuss this candidate or other critical roles, here is a link to my calendar to schedule a call:
Schedule a CallContact:
Finding the signal in the noise since 2005