AI Automation for supply chain and logistics company

Project Overview We are building an AI-powered automation solution for freight brokers, 3PLs, logistics companies, and transportation companies. We are looking for an experienced AI/automation developer to help us design and build an initial MVP that automates repetitive logistics operations involving emails, shipping documents, and Transportation Management Systems (TMS). This is NOT a project to build a new TMS from scratch. The goal is to build an AI automation layer that integrates with a company's existing systems. Initial MVP – AI Load Entry Automation The first workflow we want to build is an AI-powered Load Entry Agent. Example workflow: Customer sends a load tender / shipment request through email. The system monitors a designated inbox and identifies relevant shipment emails. AI extracts shipment information from the email and attached documents. Information may include: Customer Origin Destination Pickup date/time Delivery date/time Equipment type Weight Pallet count Commodity Reference/PO numbers Special instructions The system validates required information. If required information is missing or uncertain, the shipment should be flagged for human review rather than automatically processed. Valid shipment information is sent to the TMS through an API where available. The system records the transaction and provides an audit trail showing what information was extracted and what actions were taken. A human operator should be able to review/approve exceptions. Future Automations This initial project may expand into additional logistics automation modules, including: AI freight quote generation Shipment tracking Automated customer shipment updates Delay/exception detection BOL (Bill of Lading) processing POD (Proof of Delivery) processing Carrier invoice processing Freight invoice auditing Customer service automation Carrier onboarding Email operations automation TMS/WMS/ERP integrations We would prefer to work with someone who can continue developing these modules after a successful MVP. Technical Experience We're Looking For Experience with several of the following is preferred: OpenAI APIs / LLM applications AI agents and structured data extraction Python and/or JavaScript/TypeScript REST APIs and webhooks n8n, Make, or similar workflow automation platforms Microsoft 365/Outlook and/or Gmail integrations PDF/document processing OCR/document extraction PostgreSQL or similar databases Azure/AWS/GCP TMS/WMS/ERP integrations Authentication and API security Docker Experience specifically with logistics, freight brokerage, transportation, supply chain systems, or TMS integrations is a major plus. Important Requirements The system must be designed for reliability and production use. We do NOT want an AI system blindly performing actions when it is uncertain. The solution should include: Confidence/validation rules Human-in-the-loop approval Error handling Logging Audit history Secure credential management API retry/failure handling Ability to review AI-extracted information before submission Architecture should also allow us to support multiple customers in the future. What We Need From You When applying, please answer the following: Have you built AI workflow automation systems used in production? Have you integrated with a TMS, WMS, ERP, CRM, or similar business platform? Which ones? Have you built systems that automatically process incoming emails and attachments? What would you use to extract structured shipment information from emails, PDFs, BOLs, PODs, and other documents? How would you prevent incorrect AI-extracted information from automatically entering the TMS? Would you recommend n8n/Make or a custom Python/TypeScript backend for this project, and why? How would you architect the system so we can eventually support multiple logistics customers? Please provide examples of similar AI automation or integration projects you have built. Initial Deliverables For Phase 1, we expect: Solution architecture Working email intake workflow AI shipment data extraction Attachment/document processing Structured JSON output Data validation Human review/approval process TMS API integration or test/mock integration Logging/error handling Basic administration interface or dashboard Deployment documentation Source code Technical documentation Relationship We are interested in establishing a long-term relationship with the right developer or team. The successful developer may continue working with us to develop additional logistics automation modules and deploy/customize the platform for future clients. Please start your proposal with the words "LOGISTICS AI" so we know you have read the complete project description.