General-purpose optimization
Linear & Integer Programming
Linear & Integer Programming is the foundation of optimization — and the engine underneath the CUD optimizer. It lets AI agents solve planning, allocation, selection, and scheduling problems by finding the best solution within a set of constraints.
Unlike domain-specific solvers, this is a general-purpose optimization engine. AI agents can formulate the mathematical model, submit it through MCP, and explain the results in business terms.
Under the hood, your agent translates the natural-language description into variables, an objective, and constraints, and the solver returns the provably optimal answer — deterministic and repeatable, not a best guess. Try the example prompt below to see it yourself.
Free for testing — evaluate this solver via MCP today.
The solver supports
- Linear Programming (LP)
- Mixed-Integer Programming (MIP)
- Continuous variables
- Integer variables
- Binary yes/no decisions
- Maximization and minimization objectives
- Linear equality and inequality constraints
What can it solve?
- Budget allocation
- Portfolio selection
- Resource planning
- Capacity management
- Workforce scheduling
- Assignment and matching problems
- Production planning
- Network and logistics optimization
Example agent prompt
“We have six investment opportunities: Project A (€40,000 / expected value 90), B (€25,000 / 50), C (€35,000 / 80), D (€20,000 / 45), E (€30,000 / 70), F (€15,000 / 30). Total budget: €90,000. If Project A is selected, Project D must also be selected. Projects B and C cannot both be selected. At least 3 projects must be selected. Formulate this as a mixed-integer optimization problem, solve it, and explain why the selected portfolio is optimal.”
Have a recurring optimization problem in mind? We'll build a solver for it.