Staff Deep Learning Engineer, State Estimation (R5785)

Remote; San Francisco (CA); On-site; USD 190,000 - 290,000; Full time

Job summary

Shield AI, a defense-tech company with global reach, seeks an expert to advance Hivemind SDK state estimation and vision. You will lead deep learning for motion understanding and GPS-denied localization, bridging perception and geometric estimation across flight teams.

You will own data curation, tool selection, training workflows, and evaluative metrics, delivering reusable components for VIO and terrain-relative navigation, optimized for onboard compute and latency constraints.

Qualifications

• MS in a related field with 4+ years of work experience, or PhD with 2+ years.
• Hands-on PyTorch experience: training, debugging, and evaluation with geometric consistency.
• Strong foundations in camera models, coordinate transformations, and multi-view geometry.

Responsibilities

• Develop and evaluate models for feature detection, matching, and depth estimation.
• Combine learned representations with geometric methods for localization accuracy.
• Own data preparation, annotation needs, labeling tools, and quality checks.
• Build reproducible training workflows with experiment tracking and versioning.
• Design evaluations across lighting, viewpoints, altitude, and terrain changes.
• Analyze failures and prioritize data- and model-improvement experiments.
• Collaborate with state estimation engineers to integrate measurements into VIO/navigation.
• Profile models for onboard compute and memory, aiding deployment optimization.

Skills

PyTorch

Python

Geometry

SLAM

3D vision

Dataset versioning

Experiment tracking

Education

M.S. in Aerospace Engineering, Electrical Engineering, Robotics, Computer Science, or related field

Ph.D. preferred or 2+ years with PhD

Tools

TensorRT

ONNX

CUDA

GPU profiling

Job description

Shield AI is a venture-backed defense-tech company with the mission of protecting service members and civilians with intelligent systems. Its products include Hivemind autonomy software, V-BAT and X-BAT aircraft, and Aechelon simulation and synthetic reality technologies. With offices and facilities across the U.S., Europe, the Middle East, and Asia-Pacific, Shield AI’s technology actively supports operations worldwide. For more information, visit www.shield.ai. Follow Shield AI on LinkedIn, X, Instagram, and YouTube.

Job Description:

Join Shield AI’s Hivemind SDK State Estimation & Vision team to build deep learning capabilities that help autonomous systems understand their motion and localize in the world when GPS is unavailable or unreliable.

You will work at the intersection of deep learning, 3D computer vision, and geometric estimation, developing learned components for vision-based navigation. You will own the model development pipeline—from selecting tools, defining annotation needs, and cleaning data through training, evaluation, integration support, and deployment recommendations.

Hands‑on experience with Deep Learning and 3D computer vision and geometry is required; prior experience specifically in vision‑based navigation is optional.

What you’ll do:

• Develop and evaluate models for tasks such as feature detection and matching, visual correspondence, depth estimation, relative pose estimation, and image‑to‑map localization.

• Combine learned visual representations with geometric methods to improve localization accuracy, robustness, and recovery under challenging conditions.

• Own data preparation and supervision strategies, including dataset curation, annotation requirements, labeling tools, automated quality checks, and coverage analysis.

• Select and integrate deep learning tools and build reproducible training workflows, including experiment tracking, configuration management, and dataset and model versioning.

• Design evaluations that measure both model performance and downstream localization outcomes across changes in lighting, viewpoint, altitude, terrain, weather, and sensor characteristics.

• Analyze failures and use controlled experiments to prioritize improvements to data, supervision, models, and integration.

• Partner with state estimation engineers to integrate learned measurements and confidence estimates into VIO and terrain‑relative navigation systems.

• Profile models against onboard compute, memory, and latency constraints, and work with deployment engineers on optimization and runtime validation.

• Deliver tested, documented components and interfaces for Hivemind SDK, collaborating with software, systems, and flight test teams.

Required qualifications:

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

• Hands‑on experience designing, training, debugging, and evaluating models using PyTorch or an equivalent framework, including architecture selection, loss design, optimization, and augmentation that preserves geometric consistency.

• Strong foundations in camera models, coordinate transformations, projective geometry, and multi‑view geometry, with practical experience in one or more fields: vision‑based navigation, visual geolocation, Structure from Motion (SfM), SLAM, 3D reconstruction, depth estimation or similar fields. Expertise in every area is not required.

• Strong Python skills and experience writing maintainable, reusable software. Demonstrated ability to take a computer vision capability from problem definition and raw data through training, evaluation, and integration readiness.

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

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

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

• Ability to profile inference latency and memory use, document model interfaces and preprocessing, assess accuracy–compute tradeoffs, and advise deployment engineers on export, precision, and runtime optimization.

• Ability to communicate assumptions, experimental findings, and design tradeoffs clearly and translate research into working software.

Preferred qualifications:

• Experience with aerial imagery, geospatial data, elevation maps, or matching observations across viewpoint, lighting, season, or sensor modality

• Experience with model export, quantization, TensorRT, ONNX, or deployment on embedded compute platforms.

• Experience validating perception or robotics systems on physical platforms.

• Relevant publications, open‑source contributions, or demonstrated delivery of production computer vision systems, familiar with methods including:

  • feature correlation, correlation or cost volumes, and matching methods for correspondence, stereo, optical flow, or localization.

  • learning priors over scene geometry, depth, motion, or appearance to improve estimation under sparse, ambiguous, or degraded observations.

  • applying diffusion models or flow matching to computer vision, geometric inference, or conditional generation.

• Experience in aerospace and / or defense industry.

You do not need every nice‑to‑have to succeed in this role. We welcome engineers with strong Deep Learning and 3D vision foundations who are excited to develop expertise in state estimation and navigation to apply.