Design decisions that shape every part of the platform.
Freight operations software makes tradeoffs. Here are ours — stated plainly, with the reasoning behind each one.
01
Human in the loop — always
Agents do the work. Humans approve the commitments. A rate confirmation does not leave the building until a dispatcher reviews and approves it. Invoice payment does not route until a human has seen the audit result. The approval gate is not optional — it is the design.
This matters because a freight operation that runs without human checkpoints is a freight operation that cannot recover gracefully when an agent makes a mistake. Agent mistakes happen. The system is built assuming they will.
02
Extraction accuracy is measured per field, not per document
A 90% document accuracy rate sounds good until you realize it means one field wrong on average on every load. LogAI measures confidence at the field level — origin city, weight, commodity, delivery date — and routes low-confidence fields to a human review queue before they enter the load record.
Silent errors in load records are the most expensive kind. A wrong delivery date that gets booked creates a service failure. A wrong weight creates an invoice dispute. Per-field confidence catches these before they cause problems.
How accuracy is measured →03
The audit trail is the source of truth
Every state transition on every load — whether driven by a human dispatcher or an automated agent — writes to the same immutable event log with an actor, a timestamp, and a reason. When a carrier disputes a detention charge or a shipper disputes a delivery date, the answer is already in the record.
Freight disputes are won or lost on documentation. A platform that does not maintain a complete, timestamped event log leaves its users to reconstruct what happened from emails and memory.
Why the audit trail matters →04
Beside the TMS — not instead of it
McLeod, MercuryGate, and Trimble TMW are integrations, not competitors. LogAI reads from and writes to existing TMS platforms. Dispatchers keep working in the tools they know. LogAI handles the layer the TMS does not: document ingestion, agent automation, invoice audit, and exception monitoring.
Replacing a TMS is a multi-year project with high risk. LogAI is built to deliver value without asking customers to replace infrastructure they depend on.
05
Freight terminology — not logistics abstraction
The platform uses freight-native language throughout: rate confirmations, BOLs, PODs, pro numbers, TONU, detention, accessorials, FSC tables. Not 'documents.' Not 'shipments.' Not 'tasks.' Freight brokerages and 3PLs operate with specific terminology for good reasons, and the software should match.
When the platform uses the same language as the industry, training time is shorter, errors are rarer, and dispatchers spend less time translating between the software and their workflows.