Software Engineer, Agentic Modeling & Simulation

Anduril Anduril · Defense · Washington, DC · AFS : Discovery Engineering : Discovery Engineering

Software Engineer to build agentic modeling and simulation systems for operational planning, integrating LLM tool use, multi-agent architectures, and simulation engines to shorten military planning cycles. Focus on backend services, scenario creation, entity tasking, and real-time analysis, with an emphasis on operator collaboration and system evaluation.

What you'd actually do

  1. Build agentic workflows that make modeling and simulation capabilities accessible through natural language, structured tools, APIs, and operator-facing product experiences.
  2. Design multi-agent architectures that model human decision-making, adversary responses, operational constraints, and plan tradeoffs across complex military operations.
  3. Integrate LLM tool use, function calling, retrieval, planning loops, evaluation hooks, and guardrails with simulation engines, physics/modeling backends, and mission planning systems.
  4. Build backend services and interfaces for scenario creation, entity tasking, simulation state retrieval, course-of-action generation, plan comparison, and real-time analysis.
  5. Partner with warfighters, model/sim experts, autonomy engineers, game/simulation engineers, and mission software teams to translate ambiguous planning workflows into reliable software.

Skills

Required

  • production software engineering
  • backend services
  • APIs
  • platforms
  • integrations
  • complex stateful systems
  • AI/ML
  • LLM-powered software
  • tool/function calling
  • RAG
  • structured outputs
  • agent orchestration
  • model evaluation
  • Python
  • Go
  • backend language
  • service boundaries
  • external systems integration
  • operators
  • warfighters
  • technical SMEs
  • ambiguous operational planning needs
  • product requirements
  • engineering requirements
  • reliability
  • safety
  • observability
  • debugging
  • AI systems
  • high-stakes environments

Nice to have

  • modeling and simulation
  • physics engines
  • wargaming tools
  • autonomy simulation
  • operations research
  • planning systems
  • human/adversary behavior modeling
  • DoD modeling and simulation tools
  • AFSIM
  • STORM
  • campaign-level simulation
  • mission-level simulation
  • course-of-action analysis
  • gRPC/protobuf
  • Kubernetes
  • containerized deployments
  • distributed systems
  • secure or air-gapped environments
  • defense
  • aerospace
  • autonomy
  • robotics
  • gaming/simulation
  • command-and-control
  • operational planning domains

What the JD emphasized

  • Strong production software engineering experience building backend services, APIs, platforms, or integrations around complex stateful systems.
  • Hands-on experience building AI/ML or LLM-powered software, ideally including tool/function calling, RAG, structured outputs, agent orchestration, MCP-style interfaces, or model evaluation.
  • Ability to work directly with operators, warfighters, and technical SMEs to turn ambiguous operational planning needs into concrete product and engineering requirements.
  • Strong judgment around reliability, safety, observability, and debugging for AI systems deployed in high-stakes environments.
  • Experience shipping production agentic systems, not only prototypes.

Other signals

  • agentic workflows
  • multi-agent architectures
  • LLM tool use
  • simulation engines
  • backend services
  • scenario creation
  • entity tasking
  • course-of-action generation
  • plan comparison
  • real-time analysis
  • warfighters
  • model/sim experts
  • autonomy engineers
  • game/simulation engineers
  • mission software teams
  • observability
  • reproducibility
  • evaluation for agent behavior
  • inspectable
  • explainable
  • operationally useful
  • DoD modeling and simulation tools
  • campaign-level simulation
  • mission-level simulation
  • wargaming workflows
  • AFSIM
  • STORM