Microsoft
Solve allocation problems under capacity and resource constraints.
In ReviewBuilding the agent isn't a development project — it's four decisions. Here's exactly how it works.
Pick the platform you already use, add your costs, capacities and limits, and connect our MCP. After that, the agent returns the optimal plan for your problem, shows which limit is holding you back and what it costs, and explains clearly when no plan is possible — not a spreadsheet you have to adjust by hand.
Pick the platform you already use, or the one your team is standardizing on. Common choices:
You're not locked in — the same underlying intelligence connects to whichever builder your organization runs on.
Every agent builder needs to know where your optimization inputs live. There are two ways to connect it:
Add your products or jobs, profit or cost figures, resource usage, capacities, and operating limits in a spreadsheet. Use the numbers your team already has. It's a straightforward way to prepare the optimization problem without writing a mathematical model.
Connect through a standard protocol built for secure, structured access. The agent queries live optimization tools instead of relying on a file that went stale the moment you exported it.
MCP (Model Context Protocol) is an open standard that lets an AI agent connect directly to a live source — instead of you exporting a file and hoping it is still accurate by the time someone reads it.
In plain terms:
The result: your agent can request a calculated optimization plan and its supporting reasoning through a consistent connection.
A spreadsheet can help you prepare the inputs, but MCP is the preferable path when available — it is the secure, standardized way for an agent to access optimization tools and obtain a calculated plan, rather than relying on a static file that may become outdated after export.
This is the step that turns a generic agent builder into an Optimization Solver Agent. Connect the Optimization Solver MCP once, then ask your agent to solve planning, scheduling, and resource-allocation problems using the tools it exposes.
Pick your platform below. Connecting our MCP is the same idea on each one — you add it once, then sign in with your Saifs AI account.
Launch Claude Desktop. In the top-left corner, click the three-dot menu to open the app's main menu.
From the three-dot menu, hover over File. Click Settings (or press Ctrl + Comma).
In Settings, select the Developer tab. Click Edit Config at the bottom of the panel. A File Explorer window will open at the Claude config folder. Click to select claude_desktop_config.json.
With the config file open, add this server — not the generic Saifs AI MCP. This one is Optimization Solver only. Then restart Claude.
Open the Cursor application on your desktop. On the home screen, choose Open project, Clone repo, or Connect via SSH to begin working on a project.
Click the gear icon (top right) or go to File → Settings. In the left panel, click Tools & Integrations. Scroll to MCP Tools and click Add Custom MCP.
After clicking Add Custom MCP, Cursor opens the MCP config file. Add this server — not the generic Saifs AI MCP. This one is Optimization Solver only. Click Connect next to the new MCP server under MCP Tools to start the integration.
When prompted, select Open so Cursor can process the URI and start the login flow. Sign in with your Saifs AI account, then click Open to finish the connection and confirm the toggle is ON.
From the ChatGPT home page, go to Settings.
Click Connectors in the settings menu. Click Create in the browse connectors section.
Enter this server — not the generic Saifs AI MCP. This one is Optimization Solver only. Then click Save.
Click Save. You can now ask for the best production mix, budget allocation, and job schedules in ChatGPT.
Open the VS Code marketplace and install an MCP-compatible client.
Open the command palette, type MCP: add, and select MCP: Add Server to start the wizard. Search MCP: Add Server and choose HTTP (HTTP or Server-Sent Events) to connect to a remote MCP server.
Input this server URL — not the generic Saifs AI MCP. This one is Optimization Solver only. Press Enter to confirm. When prompted, choose Global (all workspaces) or Workspace (this workspace only).
Allow the server to authenticate so VS Code can talk to the Optimization Solver MCP. The MCP server should show as Running. That confirms the Optimization Solver Agent is connected in VS Code.
Once connected, our AI Agent isn't answering from a static spreadsheet — it's querying live, calculated intelligence every time. Here's what that looks like in practice.
Say you run a small manufacturing plant that makes two products, A and B. You fill in the production template with the numbers you already have: the profit on each product, the machine hours and materials each one uses, and how much of each resource you have this week. Every time those numbers change, without anyone rebuilding a spreadsheet, the MCP solves the problem again:
The same approach works for blends, shipping plans and schedules. Pick the template that matches your problem, fill in the table, and ask. That's the shift: from adjusting numbers until they look about right, to an exact plan with the reasoning shown — so you know exactly what your bottleneck is costing you.
“Find the best production mix within capacity”
“Allocate budget across projects”
“Schedule jobs on machines”
Same input, same output, every time. The optimization results are calculated by solver algorithms, not generated as a plausible-sounding guess.
Your optimization inputs and business constraints are handled through the connected MCP workflow. Use the deployment and data-handling controls configured for your environment.
Purpose-built optimization tools do the mathematical work instead of an LLM re-processing every constraint and calculation on every query.
The workflow uses established optimization methods such as linear programming and mixed-integer programming, rather than relying on an LLM to guess a feasible plan.
Solve allocation problems under capacity and resource constraints.
In Review
Solve allocation problems under capacity and resource constraints.
Coming Soon
Solve allocation problems under capacity and resource constraints.
Coming Soon
Solve allocation problems under capacity and resource constraints.
Coming Soon
Solve allocation problems under capacity and resource constraints.
Coming Soon
Solve allocation problems under capacity and resource constraints.
Coming Soon