A mid-size freight broker in Houston was running operations the way most brokers do: WhatsApp groups for carrier availability, spreadsheets for quoting, phone calls for negotiation, and a monthly argument about commission calculations.
It worked. Barely. And it consumed about 15 hours per week of pure administrative overhead.
Here's what changed when we connected the operation with AI agents.
The Before: Five Disconnected Processes
Quoting (3 hrs/week): Shipper sends a WhatsApp message. Someone reads it, opens a spreadsheet, looks up rates, types a quote, sends it back. Average response time: 45 minutes. After hours: next morning.
Carrier matching (4 hrs/week): Check the WhatsApp group. Call three carriers. Wait for callbacks. Compare availability. Confirm. Update the spreadsheet.
Rate negotiation (3 hrs/week): Carrier counters. Dispatcher checks margin ceiling. Goes back and forth 2-3 rounds. Sometimes accepts a deal that kills the margin because they lost track of the floor price.
Commission tracking (3 hrs/week): Month-end: pull all loads, calculate revenue minus cost, apply commission rules, handle disputes when numbers don't match expectations.
Carrier scoring (2 hrs/week): Which carriers showed up on time? Which ones had cargo issues? Nobody tracks this systematically — it's tribal knowledge in the dispatcher's head.
The After: One Connected System
We deployed five agents that share context and coordinate automatically:
Agent 1 — The Quoter: Shipper messages via WhatsApp → instant response, guided data collection, formal quote generated in under 60 seconds. Works at 3 AM. Works on weekends.
Agent 2 — The Finder: When a quote is confirmed, this agent automatically identifies the top 3 available carriers based on location, equipment type, and performance score. No phone trees.
Agent 3 — The Negotiator: Contacts carriers with offers. Handles counteroffers. Closes deals within pre-set margin parameters. The system never accepts a rate below the floor price — that math error is eliminated by design.
Agent 4 — The Accountant: Every closed load calculates profit in real time. Commissions apply rules the broker defined. The team sees their numbers live — no month-end surprises.
Agent 5 — The Evaluator: After each trip, the carrier gets a performance score update: punctuality, compliance, rate acceptance, cargo condition. Bad carriers get flagged automatically. Best carriers get priority on the next load.
The Key: They Work Together
This isn't five separate tools. Agent 1 captures the shipment → Agent 2 finds the carrier → Agent 3 negotiates the rate → Agent 4 tracks the money → Agent 5 scores the performance. Data flows automatically. No one copies anything between systems.
The Numbers
- Quote response time: 45 minutes → 60 seconds
- Carrier matching: 4 hours/week → automated
- Commission disputes: Monthly → zero (real-time transparency)
- Administrative overhead: 15 hours/week → under 2 hours (review and exceptions only)
What Didn't Change
The dispatchers still work the same way. The shippers still message via WhatsApp. The carriers still get calls. The process looks the same from the outside. The difference is invisible to everyone except the people who used to do the manual work — and the owner who sees the margin improvement.
That's the point. AI that works inside your operations, not on top of them.
Want to see what this looks like for your brokerage? Book a free 30-min strategy call at vortexagents.ai.
— Daniel Orozco, Founder — Vortex AI Agents, The Woodlands TX