Automate Carrier & Vendor Management with Custom AI

Many small logistics teams rely on their Transportation Management System (TMS) like MercuryGate or AscendTMS. These platforms are effective for booking loads but fall short in managing dynamic carrier relationships. Their carrier management modules are often rigid. If a new carrier requires a specific insurance certificate, you cannot add that as a required, expiring field in the TMS. The system often just sees a generic file attachment, forcing your team to track critical dates manually in spreadsheets.

For example, a 10-person freight brokerage might onboard five new carriers a week. Each onboarding requires checking FMCSA authority, verifying insurance documents (COI, W-9), and collecting a signed contract. A coordinator spends 45 minutes per carrier cross-referencing the FMCSA SAFER database, manually reading PDF certificates to find expiry dates, and chasing signatures via email. A typo in an MC number leads to checking the wrong carrier, and an overlooked insurance expiry date a month later puts a high-value load at risk.

The structural problem is that TMS platforms are built for transaction management, not compliance management. Their data models are fixed around loads, rates, and invoices. These systems are not designed to be extensible records for unstructured data like insurance certificates or carrier contracts. They lack the AI capabilities to parse a PDF or the workflow flexibility to handle a multi-step onboarding process that varies by carrier type.

The engagement would begin with an audit of your top 10-15 carriers. Syntora would document how each partner sends rate confirmations, invoices, and insurance updates: API, EDI, email with PDF attachments, or a web portal. This audit defines the data extraction strategy for each partner and identifies which processes offer the fastest path to automation.

The technical approach would center on a Python service using the Claude API for document intelligence. When a carrier emails a new insurance certificate, an AWS Lambda function would trigger, passing the PDF to Claude. The API would parse the document, extracting the policy number, coverage amounts, and expiration dates with over 99% accuracy. This structured data would be stored in a Supabase database, and a FastAPI endpoint would expose it to your existing TMS or a simple dashboard.

The delivered system is a fully automated process that monitors an inbox for carrier documents. It would update a central dashboard showing carrier compliance status, flagging any with insurance expiring within 30 days. You receive the complete Python source code and a runbook, and the system runs in your AWS account for less than $50 per month. Verifying a new carrier would be reduced from 45 minutes of manual work to under 60 seconds of automated processing.