Forward AI – Config Audit and Compliance with Forward Networks
Forward AI aims to revolutionize network configuration audit and compliance, particularly for organizations in regulated industries that dread annual audits. These audits are typically manual, time-consuming, and error-prone, and carry a significant risk of penalties. The traditional approach involves painstakingly listing all devices, understanding vendor-specific configuration syntaxes for each operating system, extracting data, correlating it with standards, and generating reports – a task that can span days and requires specialized expertise across multiple vendors. This complexity underscores the critical need for automation and simplification to achieve and demonstrate compliance.
Forward Networks addresses this challenge with Forward AI, allowing users to express their audit goals in natural language, such as validating consistent NTP server configurations across all devices. An agentic system then kicks in, generating a precise query to extract relevant configuration data from Forward Networks’ normalized data model, which contains configurations from all network devices. Crucially, Forward AI understands the nuances of multi-vendor environments and automatically generates platform-specific configuration templates for devices from Cisco, Arista, Juniper, Fortinet, and Palo Alto, enabling accurate interpretation and analysis of NTP settings.
This automated process swiftly assesses hundreds of devices, providing a comprehensive report detailing device names, types, configured NTP servers, and compliance status against the specified standard. For instance, the demo showed an audit of 124 devices completed rapidly, identifying discrepancies and highlighting specific device classes where the target NTP server was absent. This not only streamlines the audit process but also provides solid, verifiable evidence for compliance resolution, dramatically reducing manual effort, improving accuracy, and ensuring organizations can efficiently meet their regulatory obligations.
Presented by Nikhil Handigol, Chief AI Officer, Forward Networks. Recorded live at AI Infrastructure Field Day in Santa Clara on January 29th, 2026. Watch the entire presentation at https://techfieldday.com/appearance/forward-networks-presents-at-ai-infrastructure-field-day/ or visit https://techfieldday.com/event/aiifd4/ or https://www.forwardnetworks.com/ for more information.
Transcript
This is a different, uh, scenario. Here we are talking about configuration audit, like we are in the compliance team now, and this one is pretty close to my heart because a lot of our customers belong to regulated industries and they go through these annual audits, uh, to show that their infrastructure is compliant with standards. And it's a time of the year that they absolutely hate because they have to go through these like, long list of audit items, and it's like super manual, very time consuming.
And the consequences of delays or errors on that can be really bad. It can lead to hefty penalties. Let's, uh, let's jump to, uh, that demo.
Now I'm gonna start a new conversation. And this time we are, uh, we are doing audits and usually these, uh, audits have multiple audit items. Uh, we are dealing with one such item here.
I wanna validate that all my devices have, uh, the, a consistent NTP server configured like this, like one particular ip. This is one audit i I out of many that I have to go through typically. And so now let's imagine, let's, let's look at what this would entail in the, in the traditional world, like I have a multi-vendor environment.
I have devices from many different vendors, and each vendor has a slightly different syntax, a slightly different way of configuring NTP. Uh, so if, if I have to go through this audit item, then first I need to just like, make a full list of all the devices and based on what operating system is running on the device, I have to figure out, look for that specific configuration syntax, extract NTP information that is configured, and run this correlation across like what the standard is and generate this report super manual process that can easily take days. I mean, I may not even be an expert in all of those vendors.
So if I don't know what, uh, a Juniper syntax for NTP looks like, I may have to pull in a Juniper expert. Mm-hmm. Uh, and like pull in people, do this work.
Like it can, like you can, you can see why this is so painful and why, why it's such a hated thing in compliance. Let's, let's ask forward AI to do that instead. Now I've just expressed my goal, my intent in natural language, and, uh, and as I express this, there's an agentic system.
It kicks in and it starts working through. And so what it's trying to do here is it's generating, remember the data model that I had mentioned, the normalized data model? And the data model also has all of the configurations of all of the devices, right?
But what it is now doing is it's generating a query to extract the relevant piece of configuration from the data model. It's Generating, and by devices, it's pulling out the network devices, not the end points. Exactly.
It's network devices. Yeah. Right?
And what it's trying to do here, let's read what it's trying to do. So this is the kind of query that it is trying to generate, list all devices in the network, uh, and their NTP servers, uh, show which devices have NTP server configured and which do not. And I want to give me more information around device name, device type, and all of the n TB servers that are actually configured.
And so it's, it's, it's building a query, executing that query, and then it's trying to produce a result based on that. It's, it's done with its job and, uh, it gives me, what is it telling me? Uh, so the validation is completed across 1 24 devices.
And, uh, it is saying configuration. The target server is present on majority of production devices, but absent on certain device classes, it's giving me like, okay, it's telling me that it's not, not everything's good. Uh, so I need to go fix, I have some work to do.
Uh, but I wanna draw your attention to this. This is the NQE query that it generated. I want to, uh, I wanna show you a, a few important bits of this, but first, like in the query, you can see that it is generating configuration templates that correspond to how N TP is configured on different platforms.
This is the template for configuring NTP on Cisco ATA devices. Here's a template for Juniper. Here's a template for Fortinet.
And Palo Alto has a different kind of configuration syntax. So it's taking, it's generating templates for various platforms, and then it is running through all the devices in my environment. And based on the operating system that those devices are running, it's matching against the corresponding device template, extracting NTP information.
And then it is running this logic, uh, if, if, uh, and is checking if the NTP server, uh, the target NTP server is present in configured NTP servers. And based on that, it is, it generates this comprehensive report of all the devices in the network, what type of device it is, what are the configured N TP servers, and whether they have the target n TP server or not. And that's how it produces my, that's how it comes back with this response.
I don't have to just, like take the response. I have solid evidence that I can use to help resolve my audit item.
