Food and beverage product development moves fast. Ingredients change, costs shift, new nutrition, quality, and regulatory requirements emerge, and every decision can impact sourcing, compliance, product performance, and speed to market. TraceGain's AI formulation helps food scientists evaluate more options, compare trade offs, and move from idea to viable compliant formulas faster. Built specifically for food and beverage product development, it combines food science intelligence, supplier data, nutritional analysis, and formulation workflows in one connected workspace. Really, the goal of this tool is to accelerate product ideation. Whether that's making a new recipe, finding a new ingredient, swapping in an ingredient for one you can no longer source, those are all wonderful use cases. Now what makes this tool different than, say, going over to Copilot or Claude and saying, can you quickly whip up a recipe? Well, those tools could potentially do that if you're trying to make a weeknight meal. What Tracecan does is quite a bit different. We've designed our system to be wired up to data that is trustworthy, and this means a few things. For one, we've essentially built in food science textbooks, databases of information that aren't just listing of ingredients and their amounts, but actually how ingredients behave, including through different types of production processes. We don't just wanna develop random seasoning. We want one that has the right moisture value to adhere to a snack. Part of this, of course, is having underlying ingredient and specification data. We pull that through for your existing TraceScan's implementation where it's likely housed in the supplier management module. We also back that up with the ingredients that are part of the TraceScan's network in our database of nearly half a million. And last but not least, we also pull in USDA nutritionals for more commodity type ingredients, which can be a great starting point to formulate. Let's look at an actual example. I'm gonna go into a conversation I had earlier that I actually turned into a formula project for a spicy salt blend. First, I started with the most basic of queries. Can you make a salt forward spice blend and also make it spicy for me? The recipe agent got to work and pulled together the canvas at right and explained its work. I won't read the whole paragraph, but for example, it layered chilly heat smoky death with a light acidity for brightness. It also gave some suggestions of what we might wanna do next. In this case, though, I said, alright. This looks like a good recipe. I see that you use some ingredients out of our databases. You also use some ingredients out of our library, ones that we're already sourcing. But I wanted the system to then go find a real supplier of pepper, red crushed quarter inch in our marketplace. It did so in an instant, pulled together actually a couple 100 different results, and offered to narrow them down based off of certifications. Maybe I wanted an organic red pepper. In this case here, I kept it simple and swapped in one that I saw. The system did that and said, yeah. We do believe this is going to work. The quarter inch flakes are gonna be perfect for the recipe we're developing based off of your constraints. From there, you can keep going as far as you would like. For example, capping the amount of sodium, putting in other constraints, or perhaps even applying some broader company policies. You can set up the system to only develop products that your production facilities could make, and that is a huge part of the system, not just saying what ingredients, but determining what would be the method for them to be combined. Policies can also include things like banned ingredients, maybe you're developing for a clean label product. If you set those rules once, you could have any project from thereon adhere to them. But in this example here, I'm pretty satisfied. I got this 10 ingredient blend. I know how much sodium it will contain per 100 grams. I even have an estimate of the cost, and I can narrow down to see what's driving it. Apparently, the maltodextrin costs quite a bit. Now while developing a recipe is the big finish line of this type of tool, we give you quite a few extra abilities sitting over here as icons at right to guide that project. Those include the ability to develop a brief and kind of track your project from various stages, setting composition and nutrition goals. As you make any changes, everything, of course, is recorded. You can do a version history and pop back to any snapshot as you work on things. You also may be recording recipe metadata. What is this called? What's our goal release date? Considering costs, allergens, maybe even actually making a batch of this and doing some sensory testing and recording those results. The grand finish line of this, of course, is connecting it to your true system of record. For example, at the end of this project, I could push my finished recipe over into the TraceGain's formula management tool, which would be your end system of record. And, of course, opposite is true as well. If I have a recipe I'm storing in formula management, I could always pull that in to TraceGain's AI and adjust it as time goes by. That was just a brief tour that only scratched the surface of some of the capabilities, but I hope you found that informative, and we look forward to having further conversations.