Staff AI Research Engineer, Perception

RoboForce•Milpitas, CA

2d•Onsite

About The Position

RoboForce is an AI robotics company developing Physical AI–powered Robo-Labor for dull, dirty, and dangerous work. The company's robots are engineered for demanding industrial environments, with a focus on real-world deployment and scalability. We are seeking a Staff AI Research Engineer, Perception to lead robot perception capabilities from research to real-world deployment. In this role, you will develop and ship production-grade perception models — spanning object detection, pose estimation, depth estimation, and scene understanding — directly onto RoboForce robotic platforms operating in demanding industrial environments.

Requirements

Master's degree in Machine Learning, AI, Robotics, or a related field with 4+ years of industry experience, or a PhD degree.
Expertise in modern computer vision architectures and perception techniques including detection, segmentation, pose estimation, and depth estimation.
Proven experience deploying perception models to production systems — including model optimization, quantization, and inference runtime engineering (e.g., TensorRT, ONNX Runtime).
Experience with deep learning frameworks such as PyTorch, JAX, or TensorFlow, and strong proficiency in Python and C++.
Strong understanding of linear algebra, projective geometry, probabilistic theory, and numerical optimization, with practical experience implementing these in real systems.
Requires 5 days/week in-office collaboration with the teams.

Nice To Haves

Experience deploying perception models on embedded or edge hardware (e.g., NVIDIA Jetson, Orin) in robotics or autonomous systems.
Familiarity with 3D perception techniques — point cloud processing, neural rendering, 3D reconstruction, or NeRF/Gaussian Splatting.
Proven ability to scale perception training across GPU clusters and design robust data pipelines for large-scale visual datasets.
Publication at top computer vision or robotics conferences (CVPR, ICCV, ECCV, ICRA, CoRL).

Responsibilities

Develop and own state-of-the-art perception models for object detection, 6-DoF pose estimation, depth estimation, multi-object tracking, and multi-sensor fusion from research through to on-robot deployment.
Build and optimize full perception stacks end-to-end — from data preparation and model training to quantization, TensorRT conversion, and inference optimization on robot compute hardware.
Implement multi-task learning pipelines and distributed training workflows to maximize model efficiency and generalization across diverse industrial scenarios.
Profile and optimize deployed perception models for latency, throughput, and memory constraints on embedded robot hardware.
Collaborate with manipulation and foundation model teams to integrate perception outputs — object poses, depth maps, scene graphs — into downstream robot control policies.