Software Engineer, Perception

Google; Singapore; On-site; SGD 120,000 - 180,000; Full time

Job summary

Google is seeking a Software Engineer in the Perception team to transform AI research into robust perception for industrial robots. You will develop 3D computer vision models for object detection, pose estimation, and reconstruction, and deploy them for reliable grasping in real-world environments.

You will enhance perception software, work with PyTorch/TensorFlow/JAX, and code primarily in Python or C++. The role demands a 4-year technical degree and 1–2 years in AI/robotics.

Qualifications

• 4 year degree in Computer Science, Electrical Engineering, Robotics, or a closely related field.
• 1–2 years in AI, machine learning, computer vision, or robotics with deep learning focus.
• Experience deploying CV models: detection, 6DoF pose estimation, or 3D reconstruction.
• Proficiency in Python or C++ and modern DL frameworks.

Responsibilities

• Develop and implement robust deep learning models for 3D computer vision tasks.
• Optimize and deploy algorithms for grasping and manipulation in robotics.
• Enhance perception capabilities for dynamic manufacturing environments.
• Contribute to perception software design, implementation and docs.

Skills

Python/C++

Deep learning

Computer vision

3D reconstruction

Transformer models

Education

4-year degree in CS/EE/Robotics

Tools

PyTorch

TensorFlow

JAX

Job description

As a Software Engineer in the Perception team, you will apply your strong skills in software engineering, deep learning, 3D computer vision to implement, optimize and deploy robust and reliable perception capabilities for industrial robots. You will be instrumental in developing and deploying advanced AI models for object detection, object pose estimation, 3D reconstruction, and grasping. Your work will enable robots to intelligently interact with their environment in real-world industrial settings, translating cutting-edge deep learning research into robust, deployable solutions that make industrial robotics more intelligent and accessible.

How your work moves the mission forward

• Develop and implement robust deep learning models for 3D computer vision tasks, including object detection, object pose estimation, and 3D reconstruction.
• Optimize and deploy algorithms for reliable grasping and manipulation in complex industrial robotics environments.
• Enhance robot perception capabilities, enabling more complex and reliable interactions in dynamic manufacturing settings.
• Contribute to the design, implementation and documentation of core perception software skills within Intrinsic's platform.
• Support users in applying perception skills in production applications.

Skills you will need to be successful

• A 4 year degree in Computer Science, Electrical Engineering, Robotics, or a closely related technical field.
• 1-2 years of experience in AI, machine learning, computer vision, or robotics, with a strong focus on deep learning, including transformer based models.
• Demonstrated experience with deploying computer vision, along with at least one of the following topics: object detection, 6dof pose estimation, or 3D reconstruction.
• Proficiency in at least one modern deep learning framework (e.g., PyTorch, TensorFlow, JAX).
• Strong programming skills in Python or C++.

Skills that will differentiate your candidacy

• Expertise in computer vision and robotics.
• Experience with GPU kernel programming in CUDA.
• Experience with optimizing or quantizing ML models for deployment.
• Experience developing algorithms for robotic perception, grasping and/or manipulation.