SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salary: USD 191,000 - 253,000 / annual
Anduril Industries is a defense technology company building AI-powered military systems. The Maritime Digital Production (MDP) team develops the full technology stack powering Anduril's shipbuilding factories, including data infrastructure, manufacturing execution systems, scheduling engines, and AI systems that eliminate manual work from the shop floor and business operations.
As a Senior Numerical Optimization Engineer, you will own the scheduling optimization core that decides when and how ships get built. This role involves representing a shipyard's work as a graph of 100,000+ manufacturing operations with precedence, spatial, resource, and qualification constraints, then continuously replanning as the factory changes. You will formulate the problem, build and benchmark solvers, and keep them fast and trustworthy enough that new plans can reach the floor within minutes of a disruption.
Key responsibilities include:
- Formulating shipyard production scheduling as a mathematical optimization problem covering precedence with lag, disjunctive spatial exclusion, multi-resource capacity with qualification matching, calendars, shifts, and material availability
- Owning the solver-agnostic scheduling interface, integrating commercial and open-source solvers, and benchmarking them on makespan, stability, resource utilization, and solve time
- Building a two-tier replan path: fast local repair returning feasible schedules in seconds, and background global re-solve running for minutes
- Developing stability objectives that prevent unnecessary work movement and policies for when a better plan justifies the churn
- Establishing quality metrics when no ground truth exists: LP relaxation lower bounds, ensemble upper bounds, quality ratios, and regression tests
- Researching approaches beyond V1 solvers, including decomposition, metaheuristics, rolling-horizon methods, and non-traditional hardware
- Turning disruption events into replans by determining blast radius, scoping re-solve, merging results, and resolving boundary conflicts
- Building adaptive planning templates that generate recovery work by composing geometric queries with scheduling logic
- Composing scheduled operations into work orders for the manufacturing execution system
- Writing production-quality code with comprehensive tests, participating in code review, and instrumenting systems for observable solution quality
- Partnering with manufacturing engineers, planners, and factory operators to translate yard operations into constraints and objectives
- Working alongside computational geometry and data engineers
- Leveraging AI tooling in development workflow
- Joining an on-call rotation supporting production scheduling systems
This is a research-grade problem with production consequences. Resource-constrained project scheduling at this scale is NP-hard, there is no ground truth to measure against, and schedules must be stable enough that the shop floor is not thrashed by frequent changes. You will have latitude to develop novel approaches and the obligation to ship them.
REQUIREMENTS:
- 5+ years of experience building production software systems, ideally in fast-paced environments
- Deep expertise in numerical optimization: linear programming, mixed-integer linear programming, constraint programming, combinatorial optimization, or metaheuristics
- Demonstrated ability to take real-world problems, formulate them mathematically, and carry models into systems others depend on
- Hands-on experience with optimization solvers and libraries (Gurobi, CPLEX, OR-Tools, SCIP, PuLP, CVXPY, Pyomo) with judgment to recognize when a solver or formulation is wrong
- Expert proficiency in Python for scientific computing and robust software development, with strong foundations in NumPy, SciPy, Pandas
- Strong theoretical and practical grounding in graph algorithms, constraint satisfaction, and computational complexity analysis
- Experience with scheduling or resource allocation at scale where exact methods stop working, with judgment to know when to stop optimizing
- Experience building systems that operate reliably under real operational constraints: high availability, latency budgets, degraded inputs
- Excellent written and verbal communication, including ability to explain optimization trade-offs to non-technical stakeholders
- Advanced degree in Operations Research, Applied Mathematics, Computer Science, Industrial Engineering, or related quantitative field, or equivalent depth built in practice
- U.S. Person status required (position requires access to export-controlled data)
PREFERRED QUALIFICATIONS:
- Ph.D. in Operations Research, Applied Mathematics, or closely related field with publications or production work in scheduling or large-scale optimization
- Experience with resource-constrained project scheduling (RCPSP), job shop scheduling, or precedence-constrained scheduling in production
- Experience with dynamic or online rescheduling: replanning under uncertainty, schedule stability, rolling-horizon methods
- Experience benchmarking solvers or algorithms without ground-truth optimum using dual bounds, anytime curves, or ablation studies
- Experience with metaheuristics (genetic algorithms, simulated annealing, tabu search, large neighborhood search) and decomposition methods (Benders, column generation, Lagrangian relaxation)
- Performance-critical implementation experience in C++ or Rust for optimization kernels or hot-path graph traversal
- Familiarity with graph databases or large in-memory graph representations at 100K+ nodes
- Experience in manufacturing, industrial, or OT-adjacent domains; familiarity with BOM structures, manufacturing routing, work centers, MES, ERP, PLM, or APS systems
- Exposure to traditional and non-traditional optimization hardware: GPU-accelerated solvers, HPC, quantum annealing, QUBO formulations
- Experience in hyper-growth startup-like environments balancing speed, ambiguity, and long-term system health
- Familiarity with frontier AI tooling, AI coding assistants, and AI-enabled software development workflows
- Experience in shipbuilding, aerospace, or complex discrete manufacturing
- Eligible to obtain and maintain U.S. Secret security clearance
