Staff Deep Learning Engineer, State Estimation

This is a fully remote position, open to applicants in United States.

📋 Description

• Develop and assess models for feature detection and matching, visual correspondence, depth estimation, relative pose estimation, and image-to-map localization.

• Integrate learned visual representations with geometric techniques to enhance localization precision, resilience, and recovery.

• Manage data preparation and supervisory strategies, encompassing dataset curation, annotation specifications, labeling tools, automated quality assessments, and coverage analysis.

• Choose and incorporate deep learning tools while constructing reproducible training workflows that include experiment tracking, configuration management, and dataset/model versioning.

• Design evaluations considering variations in lighting, viewpoint, altitude, terrain, weather, and sensor characteristics.

• Investigate failures and prioritize enhancements to data, supervision, models, and integration.

• Collaborate with state estimation engineers to incorporate learned measurements and confidence estimates into VIO and terrain-relative navigation systems.

• Evaluate models against onboard compute, memory, and latency limitations, supporting optimization and runtime validation.

• Provide tested and documented components and interfaces for the Hivemind SDK.

• Work alongside software, systems, and flight test teams.

⛳️ Requirements

• M.S. in Aerospace Engineering, Electrical Engineering, Robotics, Computer Science, or a related discipline; a minimum of 4+ years of relevant professional experience with an M.S. degree, or 2+ years with a Ph.D.

• Practical experience in designing, training, debugging, and evaluating models using PyTorch or a similar framework.

• Solid understanding of camera models, coordinate transformations, projective geometry, and multi-view geometry.

• Hands-on experience in one or more of the following: vision-based navigation, visual geolocation, Structure from Motion (SfM), SLAM, 3D reconstruction, depth estimation, or equivalent fields.

• Proficient Python skills and experience in crafting maintainable, reusable software.

• Experience in transitioning computer vision capabilities from problem definition and raw data through training, evaluation, and integration readiness.

• Familiarity with building pipelines for sensor data ingestion, cleaning, filtering, deduplication, and dataset versioning.

• Capability to select and integrate development tools while constructing reproducible training workflows, including configuration management, experiment tracking, checkpointing, and GPU performance troubleshooting.

• Experience in designing benchmarks, preventing data leakage, analyzing performance across various operating conditions, and connecting model metrics to downstream geometric or localization accuracy.

• Proficient in profiling inference latency and memory usage, documenting model interfaces and preprocessing, assessing accuracy-compute trade-offs, and providing advice on export, precision, and runtime optimization.

• Strong ability to clearly communicate assumptions, experimental results, and design trade-offs while translating research into functional software.

• All offers are contingent upon a cleared background check and possible reference verification.

🏝️ Benefits

• Bonus

• Benefits

• Equity

• Temporary benefits package available after 60 days of employment for temporary employees.