Landstar agents jumped between SOPs, shipment systems, and the capacity portal to answer a single customer question. RTS Labs built an AI copilot directly into the Agent Portal — RAG over Landstar's internal documents combined with live tool calls to the Shipment and Capacity APIs, deployed entirely inside Landstar's private Azure VNet. Agents now ask plain-language questions like "What's the current capacity in Lexington, KY?" and get cited answers in seconds.
Agentic AI Copilot: RAG + API Tool Use
Azure OpenAI (GPT-4o)
Azure AI Search
Promptflow
GPT-Vision
Our engineers will map your workflow and define a ship date in a 2-week Discovery Sprint.
Landstar’s agents thrive on quick, informed decisions. But the data they needed to make those decisions lived in different places — internal SOPs and reference docs in SharePoint, shipment-level detail in the TMS, and live capacity data in a separate portal. Answering a single customer question meant context-switching across two or more systems, every time.
The team didn’t need another dashboard or another search tool. They needed the answer to find them — in one place, in their existing workflow, with the same speed and precision they’d expect from a colleague who knew every system cold. And it had to live entirely inside Landstar’s private Azure environment, with no customer data crossing the perimeter.
Agents navigated two or more systems for every customer question — SOPs in SharePoint, shipment detail in the TMS, capacity in a separate portal. The context-switching tax was real and constant.
Every delay in finding an answer was a delay in serving an agent’s customer. Time spent hunting through systems was time not spent winning business or building relationships.
The team learned quickly that LLM accuracy was downstream of search accuracy. Tuning chunk size, embedding strategy, and retrieval reranking became the difference between confident answers and “I don’t know.”
RTS Labs built the copilot entirely inside Landstar’s Azure environment to meet their private-network requirements. The system combines RAG over Landstar’s internal documents with live API tool calls to the Shipment and Capacity systems — agents get one workspace, one conversational interface, and answers grounded in real data instead of stale documents. Every response includes source citations so agents can verify what they’re seeing.
Stop strategizing. Start building. Let’s map your workflow and get your AI integration into production in 90 days.
