Job description

The future of defense will be decided by those who field intelligent machines at scale. At Scout AI, we’re developing Fury, the first robotic foundation model for defense, to give U.S. forces overwhelming, adaptable, and autonomous power across every domain. Fury enables human operators to command fleets of robots through natural language, and empowers those machines to sense, decide, and act together as one. This mission will ask everything of us: urgency, precision, and relentless work. The Role We're looking for an AI Engineer to join the Fury Orchestration Team with a deep passion for developing and deploying multimodal foundation models that power autonomous teams of robots. Your work will directly influence how our system reasons, plans, remembers and coordinates across multiple robots to accomplish long-horizon missions. Depending on your background and interests, you’ll contribute across the stack: model training and evaluation, reinforcement learning, evaluation and simulation, reasoning and memory architectures, multi-agent coordination, and edge inference optimization. Expect to rapidly prototype ideas against rigorous evaluations, test ideas on real robotic mission rollouts, and break new ground as we create the first robotic foundation model for defense. We’re a startup. You’ll be moving fast, context-switching daily, and helping define the culture and process as we go. You’ll have substantial ownership, influence over technical direction, and the opportunity to help build the future of autonomous defense systems from the ground up. Responsibilities Design, train, and evaluate multimodal foundation models that enable multi-robot mission execution Drive model development decisions including model architectures, data mixtures, curriculum design, and training recipes informed by rigorous experimentation Curate and generate large-scale training datasets from both synthetic and real-world data engines Develop memory, communication, planning, and tool-use capabilities for AI agents operating across teams of robots Be obsessed with evaluation: design benchmarks, metrics, and testing methodologies that enable rapid iteration across the team Build and improve simulation environments targeted for synthetic data generation, evaluation, and reinforcement learning Translate foundational research into deployable, real-time perception and decision-making systems Collaborate across engineering, robotics, and mission teams to integrate AI systems with onboard autonomy stacks Conduct field trials and mission operations to validate system performance e under real-world constraints Qualifications 2+ years of hands-on experience building and deploying AI models, including 2+ years working with multimodal models or agentic AI systems Proficiency in Python, PyTorch, and modern machine learning infrastructure Experience training, finetuning, and evaluating large-scale deep learning models using distributed compute infrastructure Demonstrated experience improving model performance through data curation, training strategy, evaluation design, and rigorous experimentation Solid grasp of modern model training techniques including SFT, DPO, RLVR Demonstrated ability to take research from prototype to product in fast-moving environments BS, MS, or PhD in Computer Science, Engineering, Mathematics, Physics, or related technical field, or equivalent practical experience. Advanced graduate research in AI, robotics, or machine learning is a plus Bonus: Experience building closed-loop simulation, evaluation, or reinforcement learning environments for agentic or physical AI systems Must be a U.S. Person due to required access to U.S. export controlled information or facilities Passion about defense of the United States and its allies May be expected to obtain and hold a U.S. Top Secret security clearance Why Join Scout Work on the world’s most important frontier, ensuring U.S. and allied dominance in the age of intelligent machines Be a core part of a team building the first defense-specific robotic foundation model Collaborate with some of the top engineers in autonomy, AI, and national security See your work deployed on real systems Help define the future of intelligent defense systems Backed by Draper Associates, Booz Allen Ventures, and other top investors Benefits Competitive compensation package including base salary and bonus Meaningful equity Premium medical, dental, and vision plans with $0 paycheck contribution Competitive PTO and company holiday calendar Catered lunch daily and fully stocked kitchen EV charging Relocation assistance (depending on role eligibility) The stated salary range below represents an estimated base pay only and reflects consideration of multiple compensation factors. Final salary offers may differ depending on factors including, but not limited to, relevant experience or training background, specialized skills, and business needs. Most full-time positions also include highly competitive equity awards, which form part of Scout AI's overall compensation package. In addition, Scout AI provides comprehensive, top-tier benefits to full-time employees. US Salary Range $160,000 — $400,000 USD

Requirements

Must-haves

Design, train, and evaluate multimodal foundation models that enable multi-robot mission execution
Drive model development decisions including model architectures, data mixtures, curriculum design, and training recipes informed by rigorous experimentation
Curate and generate large-scale training datasets from both synthetic and real-world data engines
Develop memory, communication, planning, and tool-use capabilities for AI agents operating across teams of robots
Be obsessed with evaluation: design benchmarks, metrics, and testing methodologies that enable rapid iteration across the team
Build and improve simulation environments targeted for synthetic data generation, evaluation, and reinforcement learning
Translate foundational research into deployable, real-time perception and decision-making systems
Collaborate across engineering, robotics, and mission teams to integrate AI systems with onboard autonomy stacks
Conduct field trials and mission operations to validate system performance e under real-world constraints
2+ years of hands-on experience building and deploying AI models, including 2+ years working with multimodal models or agentic AI systems
Proficiency in Python, PyTorch, and modern machine learning infrastructure
Experience training, finetuning, and evaluating large-scale deep learning models using distributed compute infrastructure
Demonstrated experience improving model performance through data curation, training strategy, evaluation design, and rigorous experimentation
Solid grasp of modern model training techniques including SFT, DPO, RLVR
Demonstrated ability to take research from prototype to product in fast-moving environments
Must be a U.S. Person due to required access to U.S. export controlled information or facilities
Passion about defense of the United States and its allies
May be expected to obtain and hold a U.S. Top Secret security clearance
Why Join Scout
Work on the world’s most important frontier, ensuring U.S. and allied dominance in the age of intelligent machines
Be a core part of a team building the first defense-specific robotic foundation model
Collaborate with some of the top engineers in autonomy, AI, and national security

Nice-to-haves

BS, MS, or PhD in Computer Science, Engineering, Mathematics, Physics, or related technical field, or equivalent practical experience. Advanced graduate research in AI, robotics, or machine learning is a plus
Bonus: Experience building closed-loop simulation, evaluation, or reinforcement learning environments for agentic or physical AI systems

Application prep

Use the role’s language and application details while tailoring your materials.

Keywords to mirror

aigopythonmachine learning

Platform: greenhouse
Effort: quick

Benefits

Benefits detected in the listing.

Dental

Vision

Pto