Staff Engineer Hivemind Reasoning
Shield AI is a U.S. defense-technology company developing mission-autonomy software and autonomous aircraft for military and allied operations.
Funding history
Investors
About Shield AI
Founded in 2015, Shield AI builds Hivemind autonomy software and V-BAT and X-BAT aircraft for operations in contested, GPS- and communications-denied environments. Its current site also presents Aechelon synthetic-reality simulation and Vision Systems detection and tracking products.
Skills
Candidate Availability
Required and preferred rules are kept separate and reflect the wording in the original posting.
About the Role
You will design, train, adapt, and evaluate reasoning and cognition models for autonomous platforms. You will build reproducible training workflows, define data and evaluation methods, assess models in simulation and real-world data, incorporate edge constraints, diagnose failures, and communicate technical findings.
Requirements
- BS or MS in Computer Science, Electrical Engineering, Mechanical Engineering, Aerospace Engineering, or a similar field, or equivalent practical experience
- Minimum of 7 years of related experience with a Bachelor's degree, 6 years with a Master's degree, or 4 years with a PhD
- Track record of developing and deploying advanced ML models in production
- Expertise in foundation models, deep learning, model architectures, and training ecosystems
- Experience with self-supervised learning, supervised fine-tuning, imitation learning, reinforcement learning, and distillation
- Experience defining datasets, training objectives, experimental workflows, benchmarks, and reproducible comparison criteria
- Fluency in traditional AI and ML techniques
- Experience with PyTorch, JAX, TensorFlow, or similar training frameworks
- Experience with multimodal, embodied, robotic, or autonomous-system models
- Familiarity with robotics algorithms and edge-compute platforms
Responsibilities
- Design, implement, train, adapt, and evaluate reasoning and cognition models
- Validate tradeoffs among foundation models, specialized deep-learning models, and traditional AI and ML methods
- Develop capabilities for context interpretation, temporal reasoning, multi-agent coordination, and adaptation to changing information
- Build reproducible training and adaptation workflows
- Define data requirements, training objectives, benchmarks, and evaluation methods
- Evaluate generalization, robustness, uncertainty, and performance in simulation and real-world data
- Incorporate edge constraints into model architecture and training decisions
- Diagnose model failures and create feedback loops for data curation and training
- Deliver trained models, reproducible results, and performance evidence
- Communicate technical tradeoffs, findings, and model limitations
Benefits
- Bonus
- Equity