You’re Not Thinking Big Enough: Rebuilding Food Systems with AI Agents — Cody Menefee, Firecrawl
AAI Engineer
컴퓨터/소프트웨어창업/스타트업환경/생태
스크립트
00:00:00All right, hello everybody. My name is Cody. I put this picture up here because this jacket so far
00:00:20has not actually landed as well as I thought it would. No one gets the joke. So this was to really
00:00:26put it in front of your face. I don't just enjoy wearing heavily branded Letterman
00:00:30jackets. We intentionally tried to play into the bit a little bit. So my name is
00:00:35Cody. I'm on the growth team at Firecrawl and today I'm here to tell you
00:00:40you're not thinking big enough. But before I actually get into that I have to
00:00:45address a bit of an elephant in the room which is Theo stole my talk. Theo put out
00:00:50a video about a month ago called "You Need to Think Bigger" but I would like to say
00:00:54I submitted the name for this talk a month before Theo put his video out. I
00:00:58didn't steal his talk. He stole my talk. So today we're actually going to talk about
00:01:04farming and yes I actually mean farming. More specifically I mean livestock
00:01:10farming and even more specifically I mean automating pasture rotation for grass-fed
00:01:14livestock systems. I have a feeling most of you did not expect to learn about cows and
00:01:20grass and farming today but I'm here so you're going to. A little bit of
00:01:25background on who I am and why maybe you should listen to me. The short version is
00:01:30I actually have no credentials that qualify me for this talk but nonetheless I'm
00:01:34going to do my best to give it. I grew up in Kentucky. I have a background in
00:01:38blue-collar work. I was a bartender, a mechanic, a server, a whole bunch of things. Never actually a farmer though.
00:01:46And then I sort of found my way into engineering, software development, etc.
00:01:51I actually don't really like taking the title of software engineer. I'm pretty
00:01:56adverse to that. Feels like stolen valor because I am the vibe coder most of you
00:01:59all are scared of. I use AI agents all day long. I don't have any syntax memorized. I am
00:02:05not proficient in any particular coding language but I will crank out some stuff
00:02:10on a weekend. But today I actually work at Firecrawl where we're building context for AI agents. We have a
00:02:16series of web data APIs to give your agents access to the web. Again, I said I'm on the growth team,
00:02:21but today we're actually going to get into some more farming stuff. But first, a bit of credential
00:02:26I do have is this is a real picture of me hauling turkeys on top of my Tesla, and I do still have a
00:02:33crack in that glass ceiling because of it. This was in Nashville where I live in the middle of a
00:02:38residential neighborhood where you are not allowed to raise turkeys. But I raised 10 turkeys in my
00:02:43backyard because I wanted to know what it was like to actually raise livestock myself that I would eat.
00:02:49It's a very mentally difficult process, if I'm being totally honest. But this was me loading them up
00:02:56onto the roof of my Tesla, and then I drove for three hours with them to the processor, had to stop at a
00:03:01supercharger on the way, and lots of people were taking pictures. And what I can tell you is if your
00:03:07range is sufficiently decreased when you have a giant windbreak full of turkeys on top of the roof.
00:03:13So that was quite the anxious drive, I can tell you. I also almost left engineering to be a farmer. I
00:03:21really wanted to raise chickens. This is a real product image that I came up with. I wanted to wrap
00:03:26turkeys in white wrapping and literally just slap the word "e-e" on top of it. I thought it was provocative.
00:03:31I thought I would get you to buy chickens. But I realized it's actually really hard to make any money
00:03:36farming. Surprise, surprise. And I have a wife. I have two kids. It didn't feel right to ask them to
00:03:46give up the lives they had so that I could go cosplay as a farmer and raise chickens. So
00:03:51I decided to pivot and see if there were ways that we could scale farming itself and the types of
00:03:57systems that I'm interested in when it comes to livestock agriculture. So three things I want to
00:04:02get accomplished in this talk is one, convince you all to pursue bigger ideas. I think a lot of these
00:04:07talks, a lot of these conferences, a lot of us individually spend a lot of time talking about
00:04:12building software for people who build software, for people who build software, so on and so forth. And I
00:04:17really am just here to challenge you that there are other problems to solve than just another MCP
00:04:22for another SaaS solution at another company. But also I'm just really trying to take advantage of
00:04:27a captive audience. If you corner me anywhere at any time, there's a good chance I will talk to you
00:04:32about farming. So here I am. And hopefully I can convince you to come work at Firecrawl.
00:04:39So first things first, I believe livestock belongs on pasture. I think animals should live on grass.
00:04:45I think it's better for the animal, the consumer, the farmer, the ecosystem. I can give you a whole
00:04:48TED talk on each of those if I need to. You can find me later if you need me to tell you why it's
00:04:53better for animals to be on grass. But I don't have enough time to get into all of that. Take my word
00:04:57for it. Let's start there. The assumption is animals should be on grass. This is the goal I want to hit.
00:05:05I am not actually anti-containment farming. I think there's a reason we needed to do that. But 97% of cows are
00:05:11still currently finished on feedlots. 3% are raised on pasture. My opinion here, more animals could be on
00:05:17grass. I want to try to figure out how we get more animals on grass. The question is, why aren't they on
00:05:22grass? And that is, labor is the bottleneck. It is a pain in the ass to actually raise animals on grass.
00:05:28Pasture done right actually means moving animals constantly. And that takes a lot of work. If you think
00:05:35about grass-fed beef, you might think of I have 100 cows, 100 acres. I put 100 cows on 100 acres. They eat grass.
00:05:41I got beef at the end of the year. That's not quite how it works. You will very rapidly decrease the
00:05:48quality of your pasture if you just let cows graze where they want because they'll graze their favorite
00:05:51things, ignore things that they shouldn't, trample areas consistently, so on and so forth. So the solution to that is
00:05:58rotational grazing. What this means is you break up your pasture into individual paddocks where the animals
00:06:04have enough food for one day and then you move them every single day. This allows certain areas to rest
00:06:10and other areas to be grazed and over time will increase the efficacy of your pasture. But this takes
00:06:17a whole, whole, whole lot of work. This means you have to move fences, animals, water, and keep track of it
00:06:25every single day in order to appropriately move the animals as often as they need to. There are some
00:06:33solutions actually trying to work on this problem. You may have seen a company called Halter in the news
00:06:38recently. Peter Thiel invested at a two billion dollar valuation. No Fence is another company. What these
00:06:44companies do is provide collars for the animals connected to GPS satellites that allow you to draw
00:06:50virtual boundaries where you can move the animals remotely. I think this is a great step in the
00:06:56direction of trying to expand labor but this has a problem which is you have to know where to move the
00:07:05animals. This is not a science to actually be honest with you. You can't just move them in a straight line
00:07:10across the pasture routinely every single day to the same part of land. The reason is is grass doesn't
00:07:17grow the same every single day. There are drought conditions, rainfall, how much impact a particular
00:07:23section of the paddock has had and the way that this is sold today is actually farmers going out on
00:07:28pasture putting eyeballs on the grass and making intuitive decisions about where the next best move should
00:07:33be. So the question is how do we replace the farmers eyes on pasture so that they can remotely make
00:07:41educated decisions on where to move their virtual fences? There's a bit more that actually goes into
00:07:48this as well and that is you can't just you have to also know how tall the grass is. Grass has a growing
00:07:54cycle. If you grow it way too short it takes a really long time to come back. If you let it go too long
00:07:59it becomes old and bitter and the animals don't like it. There's this juvenile sweet spot that you
00:08:04want to keep the grass in. You want to cut it before it gets too tall but then you also don't want to
00:08:08cut it too short. You need to keep the animals moving and then constantly coming back to the same pasture
00:08:13so that your grass stays at the most optimal growing age and constantly has the most productivity possible.
00:08:21So a couple of ways that we can do this. These are things these are my solutions. This is something
00:08:26I've actually been working on thinking about how we can do this. A couple of options that I have are
00:08:31drone orthomosaic maps. If we could find a way to automate drone flights we could go fly them around
00:08:36our pasture take a whole bunch of pictures get some very high fidelity high resolution images of the
00:08:41grass that farmers could analyze. The problem with this is there's a lot of skilled upgrade you need to do with
00:08:48the farmers to teach them how to fly drones. A lot of regulatory issues with keeping the drones in
00:08:52sight and ideally this would be autonomous and there currently isn't a jurisdiction in the world that
00:08:56has approved autonomous drones for these types of applications. So this is a really big bottleneck.
00:09:01I think it has pretty high fidelity in the quality of imagery but is going to be a hard problem to
00:09:05solve in terms of actually getting all those hurdles accomplished. Satellites is my most favorite option
00:09:11today. There's a really cool company called Planet out there taking pictures of the entire globe every single day
00:09:17with a one by one meter resolution but there's still those satellites really high up in the sky
00:09:25and it's hard to tell some of the things you need to tell to actually make those educated decisions.
00:09:31The middle photo here is actually from a friend of mine out in Missouri working as a research grad
00:09:35assistant at the Missouri Lincoln University and this idea is just putting a trail cam next to a tree
00:09:42and some measuring apparatus that that camera can look at and just figuring out how tall is the grass in
00:09:48relation to that particular object just so that we have some sort of reference point that we can use to
00:09:54see how well the grass is growing back. If we can solve this problem along with the
00:09:59the collar situation I think there's a world here where we can drop an LLM in the middle of this loop
00:10:07and start to work on autonomous grazing operations and so what this would mean is an LLM essentially
00:10:14making the next best decision on where the animal should be any given day but this is a multivariate
00:10:19analysis. This requires the LLM to have several data inputs including where the animals are in GPS location,
00:10:25where they were yesterday, where they might go tomorrow, what the drought condition is in the area,
00:10:30how tall the grass is across the entire farm and actually it has to make this decision not just
00:10:36on a day-to-day basis but in varying degrees of relation. So where's the best next place for a
00:10:43particular cow to be but where's the best place for the herd to be in relationship to the pasture itself,
00:10:49in relationship to the farm as a whole and then more broadly the ecosystem at large. There's
00:10:55all of these components feed back into each other and if you can optimize this entire picture you
00:11:00have a more productive farm where you can actually have more animals on fewer acres which is how we
00:11:05end up actually scaling to compete with the feedlot style where you can actually have more cows on fewer grass.
00:11:12How do we solve this problem? There's a couple of components. There's three main blockers that I think
00:11:17need to exist in order for us to actually create this system. The first one is building a knowledge
00:11:21base and this is primarily what I'm working on at Firecrawl and then an open source project I have
00:11:26called OpenPasture. The idea here is a lot of the knowledge on when to move, why to move, how to move,
00:11:32the benefits of moving, etc. is all locked up in primarily YouTube videos. There's a bunch of really
00:11:38cool farmers out there. I can give you a whole bunch of channels that you can go down rabbit holes on
00:11:42of just good old guys out in Missouri, Tennessee, Kentucky trying to move their animals every single
00:11:48day telling you what they're learning, telling you what species are best for this, what lagoons are you
00:11:54want to aim for in the biodiversity and your pasture. There's a whole bunch of things that go into this
00:11:59and we need to build that knowledge base. Firecrawl is a toolkit that I use to actually
00:12:03collect this data. I'm going out scraping those YouTube videos, scraping research papers out to
00:12:08archive, building this knowledge base up and OpenPasture is the actual repository I put this
00:12:13information in to make it available to any farmer I think that might be able to use it.
00:12:18The next thing to solve is the actual visualization layer. There's a lot of components we need to know
00:12:24about the grass that the farmer is primarily getting out of the intuition from looking at the
00:12:28pasture. The two main things worth figuring out about the pasture both where the animals are where they
00:12:34should go and where you want them to be is what is the biomass, how much foliage actually is available
00:12:40for them to consume, and then long term what is the biodiversity of that particular pasture. If they
00:12:46overgraze sections too heavily they'll start to over index on different types of cool season, warm season,
00:12:52grasses, lagoons, etc. and ideally you want a really rounded really diverse pasture over time to make sure that
00:12:58the cattle are getting the nutrients they need so you don't have to supplement with things like hay, copper,
00:13:03aluminum, etc. Ideally they get all of the macronutrients and micronutrients from the grass itself which
00:13:10becomes an entirely ideally hands-off system. And then the third one is those geofence companies, so
00:13:19NoFence, Halter, while I appreciate the technology they're trying to push forward, I have a pretty strong
00:13:24disagreement with them which is in order to use their software they require you buy their collars and you
00:13:29can't plug your own software into their collars. From a business standpoint I get why this is, from an
00:13:34industry standpoint I think it's really a pain in the ass. I would like to innovate on the software layer,
00:13:39I would like to push GPS locations to these collars that my LLM can predict. I don't want to have to
00:13:45rely on their software to do this because I don't think it's as good or I think I can make it better,
00:13:49I'm being totally honest with you. So a bit of the purpose of this talk is actually a call to action
00:13:55for you all in the audience. I need someone to make me a caller. I need it to be open, the APIs need to
00:14:02be open. Ideally it's an off-the-shelf solution, some component parts that we can slap together.
00:14:08Farmers are pretty scrappy and like to heal their own things. There's a lot of analogity towards
00:14:13John Deere and this sort of like right to repair. So my ask to anyone maybe looking at this problem
00:14:19is design me a caller where the patent can be open and the APIs are open so that we can compete on
00:14:23software and optimize this solution. The next thing I'd like to maybe tease you about is this actually goes
00:14:31beyond just ruminants. So ruminants is a type of animal, cows, sheep, goats, those are all ruminant
00:14:36animals. They chew grass, they digest it in their ruminant, which is an organ further called ruminants.
00:14:43But there's actually an additional benefit we get where we can stack species on these pasture rotations.
00:14:48This is a company called Pasture Bird. They were a big catalyst for me to really get and obsessed with
00:14:53this idea. What they did is took your normal chicken house, put it up on big wheels, and automated the
00:14:59movement so it creeps its width every 24 hours across pasture. The reason you can do it this
00:15:05scientifically with chickens is because they don't actually get most of their nutrients from the
00:15:08grass. You have to supplement them with grain feed because chickens are omnivores, not quite just
00:15:13herbivores. So you can just inch this coop across the grass, giving them a fertile land to grow on. The
00:15:22nitrogen from their droppings actually help as a manure or as a fertilizer for the grass itself. And
00:15:30there's an added benefit here when you stack the ruminants with the chickens. If you run your ruminants
00:15:34first, they sort of cut off the top of the grass, and the chickens come behind them and peck out the
00:15:39parasites from their droppings. And it reduces the parasite load overall across your farm, which reduces the
00:15:45medication expense you have to actually pay to keep your animals healthy. And over term you have a more
00:15:50robust seed stock or breeding stock so that you can have stronger animals over time that require fewer
00:15:56interventions and can be left alone to just eat grass and turn into meat eventually. So the three things
00:16:04I really hope you all take away from this talk is one, I think we need to find big real world physical
00:16:09problems that we can solve that require multivariate analysis and not quite yes or no decisions. There
00:16:17are lots of problems out there that don't actually have deterministic solutions. I hear a lot of
00:16:22engineers talk about how we turn LLMs into deterministic processes. And my contention is actually
00:16:27there's a lot of problems that you can't solve with deterministic algorithms. This is one of them.
00:16:33It's a multivariate analysis. There isn't a next best paddock to move to. There's just your best
00:16:38guess on where you think they should go. And I think if we can take systems like that, these
00:16:44multi-data input systems, and drop an LLM in the center to actually reason over the data and at least
00:16:49make a suggestion that the human can confirm or deny, we can really start to scale systems like this that
00:16:55are very much restricted by the farmer's ability to scale their own labor, their own decision-making power,
00:17:02and really give them the tools that they need to grow their operations to hopefully, I think, all animals could be
00:17:08raised on grass if we solve these problems. And then the last one is, is maybe you can come help me solve
00:17:13some of these problems. The number one problem is actually just giving the agents the context they need,
00:17:19gathering the data, packaging that data, and then presenting it in a way that the LLM can reason over.
00:17:24And that's what we do over at Firecrawl. So you're not thinking big enough. Firecrawl is where we're
00:17:29building the context layer for AI, and I hope you can come build it with us. We're hiring.
00:17:34So here's all the job postings we currently have. Go to our website. Maybe find one that works out for
00:17:39you. Reach out to me. We'd love to have more people trying to figure out how we get data off the web to
00:17:44solve some more of these complex problems and present that data as context to these AI agents.
00:17:50And perhaps we can make the world a better place. My name's Cody. Open Pastures is my open source
00:17:57project. Firecrawl is where I do my day-to-day life. And these are my socials. I'll hang around
00:18:02for a little bit. I'd love to chat more about animals, cows, birds, all you like. Thank you very much for coming.
커뮤니티 글
아직 글이 없습니다. 이 영상에 대한 첫 번째 글을 작성해 보세요!
이 영상에 대해 글쓰기