Machine Learning Engineer

The SpringCube team curated the following job opportunity to help you in your job search. Explore the position below to find your next career move.

Company Overview

A technology company is bringing advanced AI and autonomous systems into the construction industry. Its team includes experienced professionals from the autonomous vehicle, logistics, and technology sectors who are focused on deploying autonomy in real-world construction environments.

The organization has developed autonomous systems for heavy construction equipment and is focused on improving job-site safety, accelerating project schedules, and supporting critical infrastructure projects. The company has raised significant funding and has achieved fully autonomous excavator deployments in construction.

The team combines construction expertise with advanced engineering capabilities to solve complex physical-world problems that cannot be fully addressed through simulation alone.

The organization is seeking a Machine Learning Engineer: Perception to develop and deploy production-grade 3D perception systems for autonomous construction equipment. The ideal candidate will have experience architecting perception systems, training models from scratch, understanding the full perception stack, and deploying machine learning solutions at scale.

The role involves working with both computer vision and LiDAR-based approaches. The successful candidate will understand that effective perception systems extend beyond individual models and will have strong intuition for data quality, model failure modes, corner-case evaluation, data pipelines, auto-labeling, and the statistical properties of perception systems.

What You’ll Do

• Design and develop early-fusion architectures that combine raw LiDAR and camera data for object detection and semantic segmentation.
• Develop and train state-of-the-art perception models, including BEV-based transformer architectures.
• Build robust perception systems capable of handling dynamic occlusion, including visibility of the robot’s own arm or bucket.
• Develop solutions that perform reliably under challenging environmental conditions such as dust, snow, rain, and high-vibration environments.
• Optimize machine learning models for real-time inference on embedded edge hardware.
• Debug system-level issues involving sensor calibration drift, latency bottlenecks, and other deployment challenges.
• Collaborate with engineering teams to develop state-of-the-art representations for downstream autonomy applications.
• Evaluate corner cases and investigate the underlying data and system factors contributing to model failures.
• Contribute to data infrastructure and pipelines to improve ground-truth quality and overall perception performance.

What We’re Looking For

• 5+ years of experience taking deep learning models from research through real-world production using PyTorch, TensorFlow, or JAX.
• Strong understanding of 3D geometry, including SE(3) transformations, homogeneous coordinates, and intrinsic and extrinsic sensor calibration.
• Ability to accurately project 3D LiDAR points onto 2D image pixels.
• Practical experience with early-fusion architectures that combine modalities at the feature level, such as BEVFusion, TransFuser, or PointPainting.
• Experience with state-of-the-art object detection using modern transformer-based architectures such as DETR and PETR.
• Experience with temporal perception models such as PETRv2 and StreamPETR.
• Advanced Python programming skills with the ability to read and write systems-level code in C++ or Rust.
• Understanding of memory management and real-time system constraints.
• Strong data intuition and an understanding of the importance of data alignment and ground-truth quality in robotics and perception systems.
• Ability to work across both machine learning models and the underlying data and infrastructure systems.

Ways to Stand Out

• Experience with voxel-based or occupancy representations, including occupancy grids, NeRFs, or voxel-based approaches for terrain mapping.
• Published research in leading conferences such as ICRA, IROS, CVPR, ECCV, ICCV, CoRL, or RSS.

Disclaimer

SpringCube curates tech job listings from various company websites to support tech professionals globally.

1. No Endorsement: Job ads on SpringCube do not imply endorsement of their authenticity or quality.
2. No Client Relationship: This company is not a client of SpringCube unless stated.
3. To Apply: Click the Apply button to be redirected to the hiring company’s application page for this job.
4. No Liability: SpringCube is not liable for inaccuracies.