Non-technical AgTech founder? How to get your first product built
Know farming but can’t code? Remi Schmaltz shares how to get your first AgTech product built and funded.

How to get your first product built?
Short answer: You don't need to write code to build a successful AgTech company. You do need someone on your side who can turn farm knowledge into working software. Pick one narrow, expensive problem. Validate it with real customers. Get a small prototype built by a technical co-founder, a venture studio or an AgTech development partner. Fund it with your own money, friends and family, grants or angels, and plan how it will be sold before you build anything big.
Remi Schmaltz knows both sides of this. He grew up on a farm outside Calgary and co-founded Decisive Farming, a precision agronomy company that TELUS later acquired. He then launched the early-stage fund Koan Capital, and today he is the founder and CEO of Brilliant Harvest, an AI customer experience platform for farm equipment dealers. By his own description, he is "a non-technical founder." In Season 1, Episode 12 of the Digital Ag Global podcast, he explained how he gets software built anyway.
This article pulls together his advice and lessons from three more Season 1 episodes. Raphael Jung of Sproot shares how his team built its first product. Susanne Cork of the Ottawa Innovation Farm and founder John Fielding talk about why startups fail. Investors Chris Rawley and Ben Williamson cover funding. The article also draws on Qaltivate's own prototype work.
Do you need a technical co-founder to start an AgTech company?
Not strictly, but you need technical leadership from the first day. Remi learned this the hard way.
At Decisive Farming, all three co-founders came from agriculture. "We had a strong agronomy [team]… but none of us were software engineers," he said. Their hardest hire was a VP of software who could bridge farm knowledge and technology. "I think we went through like four different people, so there's a lot of time, money, effort wasted, and that slowed our business growth."
His rule works in both directions:
- If you're the farm or agronomy founder, find a technical co-founder or a technical partner you trust.
- If you're the software founder, find a co-founder with "deep domain expertise." Remi explained that this brings more than product-market fit: "It's initial customers, it's relationships, maybe it's capital in the space."
Yurii Kovalchuk said the same on the S1E11 panel: "Combining ag knowledge and technology knowledge, that's kind of a silver bullet when you're trying to build an AgTech product." Most of the AgTech startups Qaltivate works with have either a founder who came from the industry, as a farmer, an agronomist or someone from poultry, or a long-time industry adviser.
What about going solo? Remi is a sole founder at Brilliant Harvest. But he has exited two companies, invested in about sixteen startups through Koan Capital, and built a strong network. For a first-time founder, he says, "it's pretty tough to go it alone… your probability of failure is probably a lot higher."
What are your options for building the first version?
Non-technical AgTech founders usually choose between five routes:
Route | Best for | Main risk | Typical speed to first version |
|---|---|---|---|
Technical co-founder | Long-term product companies | Finding the right person takes months, and equity is permanent | Slow to start, fast once in place |
Hire an in-house developer or CTO | Founders with funding and a clear roadmap | Hard to judge skills if you aren't technical, and costly mistakes (Remi's "four people") | Slow: hiring, then building |
Venture studio | AI-heavy ideas that need to be de-risked | Less control, and the studio takes equity or fees | Fast |
AgTech development partner | Founders who need a prototype or MVP now, with farm context included | You still need someone internal to own the product | Fast, weeks to a few months |
No-code tools | Testing demand with a clickable demo | Rarely survives real field conditions or scale | Very fast, but short-lived |
Remi took the venture studio route. He joined a venture studio because "the biggest gap that I realized was just the speed in which all things AI is moving." The studio built Brilliant Harvest's prototype, which helped the company "get to market faster and de-risk that," and some of the studio's team later joined Brilliant Harvest to carry the knowledge forward. He called it "maybe a new age way of having a co-founder."
A specialist AgTech development partner works in a similar way. Qaltivate works like this. The partner builds the first version, hands over clean code and documentation, and helps you hire or train the team that takes it from there. Compare the trade-offs in our guide to choosing an MVP development company.
How do you tell if a development partner can really build it?
How do you tell if a development partner can really build it?
This is the hardest part for a non-technical founder. Remi was blunt: "There's a lot of snake oil salesmen out there. It's hard to validate, especially if you're a non-technical founder like me." He solved it by working with a studio CEO he had known for years.
If you don't have that kind of relationship yet, ask any team you're considering these questions:
- Show me a prototype you built in agriculture. Ask for something running, not a slide.
- How does your app work offline or with a weak signal? Field software fails here first.
- Which parts would you not build in version one? A good partner cuts scope. A weak one adds features.
- What does the AI part cost to run per user, per month? If they can't estimate it, they haven't shipped AI at scale.
- Who owns the code, and how do you hand it over? You should be able to switch teams without starting again.
- Can I talk to a founder you've built for? Ask what went wrong, not only what went right.
Start with one narrow, expensive problem
Remi doesn't start from the technology. He starts with three questions: "What do I know, what am I good at, and where's the money?" For Brilliant Harvest, the answers were agriculture, sales, and a costly talent shortage at equipment dealerships, which struggle to hire enough service staff in rural areas. "It's really about finding that problem and validating the problem… how expensive that problem is, and then finding the right tool."
In Season 1, Episode 9, Yurii explained why being specific matters so much in agriculture: "In AgTech you should find a very specific problem and tackle it with a very specific solution that's going to work, and then once you've proven it for a small group of people, you can scale." Many early-stage founders, he said, pitch a huge total addressable market. Your first customers, though, come from a much smaller group: people "who have a very specific problem that you help with."
His guest, Raphael Jung, co-founder and CEO of Sproot, is a good example. Sproot digitizes contract farming, the seasonal agreements between agribusinesses and smallholder farmers, starting in Africa. Its first version did one narrow job. As Yurii summed it up, "We are helping outgrowers to manage their contracts better." Raphael's warning: "It's a huge space and you can do so many things. But if you don't narrow it down properly, you will for sure fail."
Two habits helped Sproot stay narrow:
Build the MVP with clients. Sproot has an office in Nairobi, so the team was close to its clients. It built its first version "in close collaboration, validating [features] with our clients and the farmers… to build something meaningful from the outset."
Write your assumptions down and test them. The team compiled "quite a huge list" of assumptions about the technology, business model and clients, then "quite rigorously started validating them," which "helps you to stay realistic."
Remi adds one more way to narrow down: choose the market where the money is. Rather than starting with broad-acre crops, which are low value and spread over a large area, he suggests starting with specialty crops, where you can "charge probably… three to four times more" in a smaller region. That lets you validate the product and get commercial traction as fast as possible before you expand.
A narrow first problem should:
- Be painful enough that someone will pay to fix it this season
- Have a clear user, such as a field officer, agronomist, dealer service manager or grain buyer
- Be testable with a handful of real customers before you raise a large round
Don't over-engineer the first version
On the S1E11 panel, Yurii described a pattern Qaltivate often sees in founders: "They are so focused on the technology and innovation itself, they are trying to over-engineer it before it needs to be done. 'We will add this and that, we will add the feature, we will integrate AI into it'… but it doesn't have any value to a client. So focus on the core features… that bring value and actually have someone pay for them."
John Fielding, President and CEO of Nano Bubble Control Systems, explained where the over-building comes from: "Entrepreneurs tend to get married to their tech." He has pivoted his target market about five times in four years, and three of those pivots came from misjudging how big the problem really was. His advice: "Learn to fail fast and move forward again, fast." And talk to the people who live with the problem, because "your friends can think it's a great idea… but until you actually talk to someone who's living it, you're going to get a rude awakening."
Remi adds the money side. AI is "a tool in the toolbox," he said, alongside software and workflow automation, and "if you're just going pure AI, it's going to be pretty tough." AI is also expensive to run. Agriculture makes that harder: "You grow a crop once a year, maybe twice a year." Startups have to "align the spend of what you're doing… to the industry that you're in, not what you think your investors that are in Silicon Valley… think that you should be doing."
Sproot shows what this restraint looks like in practice. Raphael said the team had considered an AI chat feature for farmers. They dropped it for now because of "the risk of giving wrong advice." They plan to add AI later, once they have collected enough field data and photos to train it properly.
What a lean first version looks like in practice: for a digital agronomy advisory app, Qaltivate built a working AI advisor prototype in two weeks. The team was one project manager, one AI engineer and one QA engineer. The prototype linked the client's own fertilizer and agronomy data to an AI agent using retrieval-augmented generation (RAG). It ran in a React Native mobile app where farmers could ask questions about a field, upload a photo and get a recommendation. It was enough to prove the concept and make a funding and build decision, without a year of development.
For realistic budgets at each stage, see what it costs to build an agriculture app.
Build for the field from day one
Farm software has to work where farmers are. Raphael described Sproot's users as smallholders with "half a hectare, a hectare of land, being somewhere very rural, dirt roads." So "we can't simply assume everybody has a smartphone… [or] there's mobile reception on each and every farm."
Rather than building a new app for farmers, Sproot sends weekly tasks over WhatsApp, with SMS as a fallback, in local languages. "Everybody has a mobile phone," Raphael said. For founders building for rural areas, Yurii's advice from the same episode is to use technology people already have, such as SMS and existing chat apps, to help them adopt something new.
When Sproot's field teams needed to map field boundaries with no internet connection, Qaltivate built that one feature to work fully offline on mid-range Android phones. It checks GPS accuracy at the moment of capture, so poor-quality data never enters the system.
How do you fund an AgTech MVP?
Remi lays out a clear order. "Step one is founders. Putting your own money in." Next come friends and family, because early on "the credibility of your whole business is you as an individual." Then angel investors and family offices. Only after that, venture capital. "You don't just start at VC," he said. Founders who go to VCs too early "have a hundred, two hundred, three hundred meetings" and get nowhere.
Grants can play a big part. In Season 1, Episode 4, Yurii described a common pattern, especially in Europe, where early-stage companies "go and find governmental grants or nonprofit grants where they can raise as low as fifty K or one hundred K to get a POC or MVP going," and then raise a seed round. Remi said he would back a founder whose plan is to "access some grants and… bootstrap this," as long as the founder and investors agree on that plan.
Ben Williamson, co-founder and Managing Director of Grit Road Partners, explained the stages. With a prototype or MVP, you raise a little money "to go test the market, to talk to customers." That's pre-seed. A seed round is "your first proper round of capital with institutional investors," which in AgTech is now about $1–2 million, and it comes once you have market validation.
Chris Rawley, CEO of Harvest Returns, added a warning. Grants, or "non-dilutive capital," are "a really helpful thing for startups." But governments don't always fund on "market-based conditions," so some grant-funded products never sell: "It simply is not attractive to farmers because it doesn't scratch an itch that they need." Ben summed up how investors see it: in AgTech, the problem comes first, and the "solution's not first, it's not second, might not even be third."
The lesson: use early money to answer one question fast, whether farmers will pay for this. Don't spend it building a big platform. Keep, as Yurii put it in S1E11, "a very long runway and low burn rate." In Canada, start with our overview of farm and agri-food grants. For the funding stages after that, read navigating funding stages for AgTech startups.
Prove it in the field, more than once
Test farms help founders without farm relationships prove their product works. Remi called them "a no brainer" if you're missing those relationships: "If you're not doing that, you're missing something."
Susanne Cork, Director of the Ottawa Innovation Farm at Area X.O, explained why one season of results isn't enough. Her program usually plans three-year projects: "One year of success doesn't mean anything… You have a wet year, a dry year, a really crappy weather year, a great one." She also warned against pilots that never end. Agree up front what success means: "If I can solve these three things, will you buy this?"
Build for how the product will be sold
Remi's last piece of advice changes what you should build. "Sales in agriculture is hard, super hard," he said, and there are two main paths.
Direct to farm. Decisive Farming sold farm by farm, and "it took a salesperson two years to hit stride." The reason: "it's only two crops… only two kicks at the can."
Through a channel (dealers, ag retailers, co-ops). Their staff are often "order takers" for products customers already understand. Asking a farmer "to make a new budget line item… that's consultative selling," which is very different from taking an order.
Yurii's view from S1E11 is that founders should "start selling before actually developing your technology." The founder "should personally sell it to a bunch of early adopters." When you visit farms, follow Remi's advice: be genuine, listen more than you talk, and "under promise and over deliver."
What this means for your MVP: if you'll sell through dealers, the first version may need a dealer-branded experience, simple onboarding and reporting that the dealer's team can show off. If you'll sell direct, build the proof of value into the first season, because you may only get one or two seasons to show it.
A 7-step plan for non-technical AgTech founders
- Write down the problem and its cost. Who has it, how much it costs them each season, and how they deal with it today.
- List your assumptions and talk to 10–20 potential buyers. Confirm they would pay, and how they buy: direct or through a channel.
- Choose your technical route. Co-founder, venture studio or AgTech development partner. Ask the six questions above.
- Define a one-job prototype. One user, one workflow, one measurable result.
- Fund it lean. Your own money, friends and family, grants or angels. Aim for weeks, not quarters.
- Test it in the field, over more than one season. Weak signal, gloves, glare, mid-range phones. Agree in advance what result turns a pilot into a sale.
- Decide what to build next based on what users did, not what the roadmap said. Then raise a seed round with proof.

-card-640x480.webp&w=3840&q=75&dpl=dpl_HBcf7gwmFDXJZzCbaN3Rh6LipaWB)
-card-640x480.webp&w=3840&q=75&dpl=dpl_HBcf7gwmFDXJZzCbaN3Rh6LipaWB)
-card-640x480.webp&w=3840&q=75&dpl=dpl_HBcf7gwmFDXJZzCbaN3Rh6LipaWB)


