Senior LiDAR Engineer, AiDash

• 5+ years of experience in applied machine learning or computer vision, with 3+ years focused on LiDAR or 3D data
• Proven expertise in 3D deep learning (e.g., 3D CNNs, PointNet/PointNeXt, SparseConvNet, Minkowski Engine)
• Experience building and fine-tuning transformer architectures for spatial or remote sensing applications (e.g., Swin3D, PointTransformer, GeoTransformer)
• Strong coding skills in Python and deep learning libraries (PyTorch preferred)
• Familiarity with aerial LiDAR data characteristics, including waveform/point density, elevation modeling, and coordinate systems (EPSG, UTM, etc.)
• Hands-on experience with geospatial and point cloud libraries (PDAL, Open3D, PCL)
• Understanding of GPU optimization and deployment in production (CUDA, TensorRT, TorchScript)
• Master’s or PhD in Computer Science, Remote Sensing, Geomatics, or related field

Desirable

• We are looking for a Staff Machine Learning Engineer with deep expertise in aerial LiDAR processing and 3D computer vision
• In this role, you will be responsible for designing advanced ML models - including custom CNN and Transformer architectures - for large-scale 3D point clouds and multi-modal remote sensing data
• Design, implement, and optimize deep learning models for aerial LiDAR data—using CNNs, Transformers, and hybrid architectures
• Build end-to-end pipelines for object detection, segmentation, and terrain modeling from geo-referenced point clouds
• Combine aerial LiDAR with other data sources (e.g., RGB imagery, DSM/DTM, hyperspectral) using cross-modal transformers or fusion models
• Implement efficient 3D spatial reasoning using voxel grids, sparse tensors, and attention-based architectures
• Drive experimentation, model benchmarking, and ablation studies to push accuracy and efficiency boundaries
• Collaborate with geospatial scientists and ML engineers to bring models to production via scalable APIs and cloud-native services
• Apply MLOps practices to manage datasets, monitor model drift, and automate retraining workflows