← All drills

Build a Carrier Booking Agent

What you'll be able to do

Drill 2 of the Freight & Logistics track. You build the agent that does the work of a freight broker's booking desk, turning inbound tender email into vetted, confirmed bookings and knowing exactly what to hand to a person. It builds on the dataset from drill 1 and teaches what changes when a tool call moves freight.

⌁ DHL Supply Chain runs HappyRobot AI agents over hundreds of thousands of emails and millions of voice minutes a year for appointment scheduling and driver follow-up. Brokers and 3PLs apply the same pattern to their tender inboxes, where a wrong booking costs a truck.
Start this internship
Create an account to unlock the 6 sections, the workbench, and askThili.
Begin

Sections

1. Build a Carrier Booking Agent
🔒 locked
2. 1 · Read the tender
🔒 locked
3. 2 · Qualify and price
🔒 locked
4. 3 · Find and vet a carrier
🔒 locked
5. 4 · Negotiate and book
🔒 locked
6. 5 · Escalate and measure
🔒 locked

Dig deeper

📄ReAct: Synergizing Reasoning and Acting in Language Models (Yao et al., 2022, arXiv:2210.03629)
paper
📄τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains (Yao et al., 2024, arXiv:2406.12045)
paper
📄Deal or No Deal? End-to-End Learning for Negotiation Dialogues (Lewis et al., 2017, arXiv:1706.05125)
paper
📄ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs (Qin et al., 2023, arXiv:2307.16789)
paper
🔗thili-ai/thilisynth: the data this drill runs on (tenders, carriers, rate card and ground truth)
code