Corey Fransen

Saw this on newagtalk today (posted awhile ago). Nice site that has a lot of good information and is well executed. agsist.com

I built two MCPs: one for John Deere Operations Center, one for AcreBlitz. Connect both to Claude and ask "can I spray Engenia on these fields?" Your field boundaries come from John Deere, ESA rules and label data from AcreBlitz, answer the question and Claude creates a clean report. Compliance should be this easy. Data should be this connected.

@NickHorob 750…if I lose, I blame Grok, thats what it recommended based on the current replies.

Here’s an interesting marketing experiment I’m going to attempt. I’m going to give away this Michter’s 10-year bourbon this weekend. How to enter: 1. Subscribe (or be subscribed) to my free NerdOutonBusiness.com email newsletter 2. Reply to this post with a number from 0-1,000 Tomorrow I will use a random number generator to select a number (I’ll record it). The closest person to that number who is also subscribed to my email newsletter will get this bottle of bourbon for free.

You may need to go to bed...or see a therapist if your browser tabs look like this at 10 o'clock at night.

@NickHorob @steeke7 @nfaleide haven’t looked at it closely, but appears interesting.

@coletenold @NickHorob @roboflow SAM3 model is goated!

@NickHorob Good application for SAM3 if you want to get even more accurate /automated ( @roboflow or other provider)

Claude was able to estimate the size of this pond (2.75 acres) by counting its pixels and comparing that to the number of pixels contained in the 200 ft scale line at the bottom right of this image. Each pixel is 1.42 ft. Pretty mind-blowing that it can do this analysis from a one-sentence prompt!

@JanetteJoyB @NickHorob needs to enter the chat…

Farmers, ranchers, processors -- how are you using Claude so far?

@nfaleide @NickHorob I honestly just assumed it was Nathan’s side project

I think I found the most viewed piece of AgTech software in 2026..... Perplexity releases one of the more promising tools I've seen, and as a part of their demo, they're updating an application. What is the application? An ag software tool: "Solaris - Agriculture Intelligence, From Orbit" The most popular AgTech tool of the year...a fake one 😬

Interesting how well the guidance lined up with the thread in newagtalk. 1. Pulled weather forecast for Bloomfield, MO 2. Searched extension/agronomic guides for "Valor/flumioxazin in preemergence soybeans" 3. Searched product labels

Every B2B software company is (or should be) building a "headless" version of their product. One that can be used by agents. But "headless" doesn't mean "brainless". You don't just wrap your existing APIs into an MCP server and call it a day. The companies that succeed in the agentic era are those that take a thoughtful approach to *designing* an agentic user experience (AUX). Yes, that will likely involve APIs, MCPs and CLIs. But the difference will be in the *ergonomics* of the interface. We need to figure out *how* agents actually want to use our products/platforms. Because if all they wanted to do was use them like humans do, we have "computer use" for that. I'm personally very excited about this new agentic world when it comes to B2B software. HubSpot is all-in on building the #1 agentic customer platform. Just posted this in a private Slack thread with the HubSpot exec team: Being agentic is not just about agents running *on* our platform, it's about agents *running* our platform (being able to operate it). That's how you take AI from being a simple tool to a savvy teammate.

Something to think about if you’re applying Dicamba this year. You probably already have enough runoff points. linkedin.com/pulse/esa-prac…

Interesting that Haiku models won’t use product label tools first, but does a docs search through all of the extension docs. While Sonnet and Opus will always defer to label tools first.

@Dusty3080467325 @MaxCropProtect @ISUCropNews Not arguing the phenomenon, could make a good argument for improved stand-ability in the fall and other abiotic effects beyond yield too. It’s good chemistry, but you could have called it the “Pyraclostrobin Effect”, being generic and all.

@corey_fransen @MaxCropProtect @ISUCropNews Well can you explain the phenomenon of why there’s a bump when no diseases present ??

14 ounces of quilt xl -generic 4 ounces headline -generic 6 ounces of BIFRIN -generic $10 per acre Delivered to your door Remember @ISUCropNews said it’s equally effective on all classes of diseases Don’t be fooled by BigAgChem

@Dusty3080467325 @MaxCropProtect @ISUCropNews “Headline Effect”…so branding still matters I suppose

@corey_fransen @MaxCropProtect @ISUCropNews The Headline Effect for yield bump

@Dusty3080467325 @MaxCropProtect @ISUCropNews Why not just drop the Headline altogether and save more money? Already have the group 11 from the Quilt, why add another at below label rate?

@corey_fransen @MaxCropProtect @ISUCropNews The spillover from the Quilt will have it covered If anyone wants to spike it up they are welcome to The $10 price point is most important in $4 corn environments

@Dusty3080467325 @MaxCropProtect @ISUCropNews All for generics, but why not bump the rate up to at least the labeled range for Headline? Resistance buildup has a much greater long tail cost than just using the right amount the first time.

Everyone has an opinion We are full rate on the Quilt XL The Headline when you measure the actives of carbamic acid, [2-[[[1-(4-chlorophenyl)-1H-pyrazol-3-yl]oxy]methyl]phenyl]methoxy-, methyl ester We end up getting as much as many of the new blends I get if you don’t like the economic price I post as it makes everything you sell look like you are robbing the farmer

This question from newagtalk actually surprised me, I expected it to fail given the use of acronyms and shortening of the product names. Along with Dual Magnum not being the simplest label to understand, I even struggle reading through it. Agentic file search for the win.

@calcsam @NickHorob It just didn’t end up making sense for the product. Mastra performed well with it though and it was reliable, just didn’t end up making sense.

@corey_fransen @NickHorob Curious why you didn’t release the agent?

A lot of love for what the Mastra team has built and continues to build.

Used them for my label conversion agent, using them within the aerial platform I’ve been building for all of the agentic implementations, using them in our ESA advisor portal to parse out recs and process the data into our system in a structured way, and and used it to build out an ESA compliance chat agent which I never ended up releasing. It’s been the most reliable and they helped solve a lot of observability challenges and I like their observational memory implementation.

@corey_fransen What kind of agents have you built with Mastra?