I am building a self-service freight brokerage website whose core intelligence is an automated load-matching engine. The AI should ingest available loads and truck capacity in real time, score the best matches, and surface them instantly to the right party without human intervention. Two types of users will work inside the portal—carriers and shippers—so role-based dashboards, permissions, and data views are essential. Each user needs to be able to create and manage an account, see their active and historical loads, and interact with the platform from any device. Beyond the matching logic I also need: • Seamless, PCI-compliant payment processing so shippers can pay and carriers can receive funds inside the platform. • A built-in customer support chat that ties conversations to specific loads and accounts. I am open to the stack, though something modern (React or Vue on the front end, Node.js, Python, or Go on the back end, with a container-friendly cloud deployment such as AWS or GCP) will likely make scaling easier. If you already have experience applying machine-learning frameworks—TensorFlow, PyTorch, or an equivalent—to logistics data, that will move you to the top of my list. On delivery I will review: 1. A functioning web portal with separate carrier and shipper logins. 2. A live AI model that produces automated load matches against sample data. 3. Fully integrated payments and support chat, ready for production. If you can demonstrate previous freight, supply-chain, or marketplace work, please reference it when you respond.