We’ll go ahead and say it up until now, the whole concept of applying AI technology in the logistics industry has been a little bit too abstract. Sure, “optimization through machine learning!” “Analyzing big data and predicting trends!” But then, what do you do? You still move your data around in spreadsheets, make check calls, and manually enter information.
What is unique about Alvys Foundry? Here’s the deal: it’s not just another AI service that you should find the way to integrate. It’s a full-fledged platform for developing your own AI solutions directly in Alvys TMS. These AI agents will have certain functions assigned to them according to their job description. Not just chatbots, but digital workforces for automated workflows, such as detention tracking, document intake, and load building.
Today, with thin margins in the freight business, even the smallest inefficiencies in the operation cost money. With constant pressure to work faster and more efficiently while delivering a high level of customer satisfaction, teams find themselves overwhelmed by tasks.
Now, let’s talk about the specifics of Alvys Foundry and why this “native, not bolted-on” strategy is significant.
What Are AI Agents, Really?
But before we discuss Alvys specifically, let us elaborate on the definition of an AI agent in this context.
Imagine your typical automation system is a bunch of ‘if this happens, then that will happen’ instructions. It follows the logic strictly but is incapable of handling anything that is not pre-defined. A spreadsheet will not be able to adjust itself to any unpredictable conditions.
An AI agent, however, utilizes large language models and its ability to reason in order to deal with the more sophisticated scenarios.
Here is a practical example for you to understand the difference between the two:
Automation: If there is no update on a load’s status for 2 hours, send an email notification to the dispatcher.
AI agent: If there is no update on a load’s status for 2 hours, use the information on the GPS to determine if the truck has stopped somewhere near a known shipper. If the driver is on the detention clock, then automatically record the detention time and inform the broker. If the driver is running late, recalculate the estimated time of arrival.
That’s the difference. The agent doesn’t just flag a problem; it investigates, reasons, and takes action.
AI in Alvys Foundry: Agents with a Job Description
What is so special about Alvys Foundry is not only the fact that they have AI agents but also how those agents were created and how you can work with them.
“As we say in Foundry, ‘This is not AI for show. This is AI with a job description,'” said Nick Darman, CEO at Alvys.
According to Alvys’ observations during the past year of their customers’ use of AI and where and why it worked or did not work for them, they just added more logins to the already overloaded system. The solution to the problem came in the form of Alvys Foundry, where freight operators could integrate automation and AI into their existing processes.
Three Options for Onboarding
Alvys Foundry has set up 3 ways to get on board, which is a good strategy since all businesses have different capabilities and risk appetites:
- Prebuilt agents: More than 20 prebuilt templates that cover popular processes. If your business processes are standard, then you can use these agents right away without any adjustments.
- Building your own agents: The platform provides you with the tools to build your own agents. It gives operators the ability to do this by themselves without depending on Alvys engineers.
- Building custom agents by Alvys: In case of some specific and complicated workflow, the engineers from Alvys will build agents for you.
It is important to have this flexibility since the company does not try to fit everyone into the same category but tries to meet customers’ needs.
Why “Built-In, Not Bolted-On” Matters
Now, this is where it gets a little bit technical, but I’ll explain. In order to actually do its job and provide value, an AI agent platform needs access to very detailed operational context: lane history, customer rules, documentation, margins, appointments, and exceptions. It simply cannot function in a vacuum. The majority of the TMS platforms, opening up their data to any kind of third-party agent platform, are both a security and reliability risk. You are basically giving access to your highly sensitive operational and financial information to the outside software.
The Alvys Foundry was designed internally precisely to get around this limitation. It leverages our data infrastructure, including more than 120 integrations, our native EDI capabilities connecting to hundreds of shippers, and our SOC 2-compliant secure data foundation. This also means that AI has all the context it needs. It knows lane history, customer rules, and appointment times since all of this data is already available in our system.
Data Privacy and Security
Moreover, Alvys makes enterprise agreements with the best model providers. Thus, customer data will not be used for training public models. Your operations data remains your data.
In addition, there is Agent Shield in Foundry, which is a governance layer that tracks agent activity and ensures traceability of actions and routing of high-impact decisions to human review.
What Can These Agents Actually Do?
The first thing that needs attention is the manual effort that eats up a great deal of time in the transportation business. For example:
- Automation of check calls: The agent does the job instead of a manual person calling the carrier.
- Invoice and settlement preparation: With the help of an agent, the information needed for invoicing and settlements can be prepared by the agent to minimize the gap between delivery and invoicing.
- Track-and-trace exceptions: In the case where the track-and-trace data of a shipment goes beyond the usual parameters, the agent steps in and acts accordingly.
- Detention tracking: The agents can also track the detention process.
- Intake of documents: Instead of manually handling the document, the agent extracts the necessary information from the document.
- Load building: The agents can help in building loads depending on certain parameters.
The Intelligent Model Picker
It’s worth noting one technical feature here: Alvys Foundry contains an intelligent model picker that automatically picks the most cost-effective large language model for any particular task. Why does it matter? Running sophisticated AI models is rather costly.
Putting It into Perspective: Why It Is Important for Freight
Taking it into perspective, Alvys Foundry reflects a larger trend in the way that technology is used for freight.
Moving from Burden to Benefit
TMS software has long been considered an obligatory tool, but not necessarily a competitive advantage. Alvy’s clients are experiencing tangible benefits that are flipping this logic:
- Increase in monthly loads by 28%
- Increase in revenues by 9%
- Faster accounting by 90%
- Less data entry by 80%
When customers are able to see ROI within weeks, rather than years, it is an investment and not just a cost.
The Efficiency Problem
In the freight business, fragmentation is an issue. The inability to grow without generating disorder due to disjointed systems, duplication, and inefficiency leads to high costs in terms of inefficiency.
This is where Alvys Foundry jumps in by automating tedious activities and allowing human labor to concentrate on judgment, service, and relationships. According to Spartan Carrier Group’s CEO, “It helps take the manual friction out of freight so our team can concentrate on judgment, service, and elite execution.”
All Fleet Sizes to Benefit from Accessibility
One of the more interesting advances made recently is Alvys’s statement that their TMS will now be accessible for all fleet sizes owner-operators, smaller freight companies, and large trucking companies.
Digging Deeper: The Alvys Perspective on the AI in the Logistics Trend
Alvys is not the only logistics company that utilizes AI agents. For instance, C.H. Robinson has 30-plus AI agents performing all kinds of tasks, including LTL pricing and missed pickups. Specifically, one agent at C.H. Robinson solves problems related to missed LTL pickups; it automates more than 95 percent of checks and saves more than 350 hours of manual labor per day. There is, however, a fundamental difference.
The agents at C.H. Robinson are developed by the company itself and sold as a service to the customers. Alvys Foundry, on the other hand, provides its operators with components and teaches them how to make their own agents out of them. The Alvys’ approach implies reduced vendor lock-in. One does not need the vendor for any customizations, integrations, or reports. One is able to evolve the agents alongside the evolution of one’s business.
This is quite an important difference from other systems that have locked many companies into vendor control and charges for everything.
Practical Points to Consider and Common Mistakes
When considering implementing AI agents for your freight transportation needs, keep in mind some practical points.
Start Small and Specific
The most effective implementations of AI agents begin with well-focused tasks. “Automate all of our logistics” is an automatic way to fail. ” Automate our detention tracking in our 10 most important lanes” is a much more practical first step.
AI agents built by Alvys Foundry already have pre-built templates for this; they deal with particular processes and not all processes.
Data Is Key
The most successful use cases for AI agents begin with narrowly defined issues. “Automate all our logistics” is sure to fail; “automate detention tracking in our top 10 lanes” is a more reasonable initial goal.
Pre-defined templates from Alvys Foundry are meant to facilitate this approach by concentrating on certain processes rather than attempting to cover everything at once.
The Importance of Data Quality
The performance of an AI agent depends entirely on the quality of the data it has access to. Inconsistency, incompleteness, and inaccuracy of information result in bad decision-making. It is the value of the “built-in” solution provided by Alvys Foundry, as the data is consistent.
Human Involvement Is Not Obsolete
Agent Shield, which is the governance layer responsible for tracking activity and channeling key decisions to humans, understands one very important thing: artificial intelligence agents are simply tools to help, but they are not substitutes for humans who still make decisions and deal with exceptions and relationships. The agents just do routine tasks.
What’s Next? Looking Forward
The $40 million Series B investment that Alvys received in September 2025 was designed to place artificial intelligence at the heart of its platform. Foundry is the first deliverable on that timeline.
The company plans to move ahead by delivering solutions that are enterprise-grade with AI at their heart, leveraging artificial intelligence, automation, and integration in order to strip away wasteful processes, make better decisions in real time, and scale operations and profit margins without increasing overhead. Freight is about to get interesting in the coming years. It won’t be whether freight becomes transformed with artificial intelligence but when.
Actionable Takeaways
- Begin by looking at those workflow activities that take up most of your time. Look for similarities in repetitive data input and check calls, status reports, and billing activities. They are the easiest processes to automate.
- Think about the importance of data integration. Are there different systems working independently? In such cases, the “built-in” strength of AI is even more significant than its capabilities.
- Consider your vendor policy. Do you really want to invest in something where you have little control over customization? Will you be able to create agents, or will the vendor decide everything?
- Start small; try to automate just one or two workflows. Prove the usefulness, and then move further.
- Remember to have humans involved in making important decisions. Governance is not an obstacle; it is an inevitable part of any process.
Resources and Next Steps
- Alvys Foundry: Please visit Alvys’ official website to find up-to-date documents.
- C.H. Robinson’s AI agents: The Lean AI strategy is yet another way that AI can be utilized in logistics.
- Freightwaves: Great resource for logistics technology news and articles.
- Commercial Carrier Journal: Good place to get information about trucking technology and compliance.
This blog post is purely meant for education and informative purposes. The advancements in freight technology are moving fast. It is advisable to confirm all the details from the right sources and seek the opinion of your legal team.
Explore Our AI Category. And if you are reading it up to here, leave a sweet comment to motivate us to write blog everyday.
