- The coolest and most challenging: pH, EC, and flow rate
The hardest part has been running everything on battery while maintaining accuracy and using LTE (2–4G) and not common LPWAN options like LoRa. I'm primarily a software guy, so the learning curve has been huge.
We built a company around this very problem- indoor farming (including vertical indoor farming) is pretty complex and by default it's energy hungry. In theory, indoor farming is very efficient for commercial food production though. I thought we could be the company that does all the plant biology, automation complexity magic for growers, and growers just do seedling and harvesting in a super basic mechanical setup.
We are working with growers in EU, and they are all actually profitable growing normal veggies (lettuce, kale, etc.) as usual. But whenever we talked to some of the fancy VC-backed vertical indoor farming companies, they would usually not entertain us and would always claim that they were going to build everything by themselves. Almost always, the leadership in these companies was the type that didn't know anything about plants, software, status quo of AI, etc.
> they would usually not entertain us and would always claim that they were going to build everything by themselves.
I'd assume it's an attempt to have exclusivity (and thus a shot at exponential growth) instead of targeting slow and stable growth ? From your description alone, it feels your business model is more targeted to small business than startups (which is a good thing IMO)
> I'd assume it's an attempt to have exclusivity (and thus a shot at exponential growth) instead of targeting slow and stable growth ?
My guess from the context would rather be that they think (rightly or not) that they can do it cheaper themselves or by paying someone else a fixed cost rather than giving out a profit share like the pricing model in the FAQ says:
> Our pricing model is based on the principle that we take certain percentage of the total profit our SaaS lets you drive home.
A lot of it is also that the tech/VC way of things is to try to build everything yourself, and to assume the mechanics of the problem is easy and doesn't require a lot of domain knowledge. Whereas the software required to orchestrate all of these systems and provide insight is complex and needs a lot of software. I think it's a classic case of misunderstanding how complicated everything is and assuming that because you have been given a bunch of VC money that it's somehow a mandate to build everything yourself. It's what happens most of the time when a "tech" company interacts with the physical world. Like how my Nest thermostat always learns to do the opposite of what I want at any point in time.
They've told their investors that everybody else does it wrong, to deliberately divorce their own valuation from reality. They don't want to be valued like a regular greenhouse company, they want to be valued as a cutting edge company with huge unbounded potential. Turning around and using COTS solutions undermines that narrative.
It's a SaaS + IoT + Plant Biology knowledge baked into one package. We are figuring things out on the fly as well (here's a link to one of the products https://www.hexafarms.com/main/hexaos). The aim is that the entire operations should be reduced to manual labor of handling the plants (and our software will inform you about that as well). Vertical indoor farming has been always close to my heart but at this point we address the wider space of Controlled Environment Agriculture (CEA) in general.
> How are you different from “VC-backed vertical indoor farming companies”?
Yikes! We are also going to be VC-backed soon. Went through Techstars recently. Hopefully, I'd have the humility to accept and not makes claims that go against the fundamental principles of physics, biology, and economics. Sorry but this is the best answer I could give.
Literally just got fired from a company that was just like that. Had a great idea, but wanted to build everything themselves vs just pieces of system that could differentiate us. Add onto that a weak business plan plus trying to compete with others in the greens space and it didn’t go very far. A lot of fun to work on though!
I second this. I would say that I am pretty advanced with Python (and Django) and same with JavaScript (Vue and Nuxt) and have written applications that got used by multiple users. I saw a sharp rise in my productivity when I knew enough about the frameworks. But Elixir + PhoenixLiveView + Tailwind has been life changing.
I learned Elixir for the sake of the joy of learning a new programming language and I kept playing with it for few random days over 5-6 months. Finally, I took the leap of faith and for our startup I started the switch to Elixir + LiveView with minimal JavaScript hooks and I feel a weird bliss that we are two engineer FTEs and I can add features on a daily basis. Why that's the case, I still haven't ruminated myself, but my guess is 1. I have gotten older, 2. Elixir is beautiful and productive by design-- pattern matching, everything is a process (so the dimensionality of time is not an issue at all) and the code is kind of a right balance of simplicity and complexity, and in my opinionated view, there is "one" right way of doing things. 3. Standard tooling (mix, ExUnit). They have enabled us to write really maintainable code and for our next hires, we are willing to pay for them to learn Elixir than switching to other languages. Of course this is only for true for our web app which is actually a weird beast that interfaces farms, sensors, algorithms, and humans.
Question for you and OP, do you find that using languages like Elixir makes it so you have to look for a certain kind of dev given that Elixir is a pretty niche community or are folks with no experience able to adapt with relative ease?
Not one of the OPs but can happily echo their experience/praise for PETAL and think I can answer your dev search question:
Anyone with a can-do attitude can learn Elixir. People with a “no thank you, I already know XYZ” mindset you really don’t want on your team.
Almost 20 years ago @pg wrote the “Python Paradox” post: www.paulgraham.com/pypar.html
The same holds true for Elixir (and Rust) these days. The people you want to hire are those who learn new skills/languages proactively.
However I can confidently say, from experience of having hired several Python devs/data scientists with zero Elixir experience, that smart+motivated people can learn Elixir in a couple of days and be productive within a week. See: https://www.verytechnology.com/iot-insights/elixir-for-pytho...
I'd argue the other way. Yes anyone with a can-do can learn Elixir. But there's a difference between learn and LEARN..
I work with a pile of rails devs writing code in elixir. In that I mean you can feel the rails in their code. It "tastes" like rails. A few python devs also write python with elixir syntax.
Finding people who know the language is hard. Finding people who are willing to risk their time on the language is also hard.
By risk. If I spend 2 years writing Elixir and then go looking for a job I'm looking at getting a job with a language that's not popular. So my pool of place that will even look at me is smaller. Hiring managers/gate keepers might not see..
After 12 years of c++ I got a job where I needed to code in Delphi-Script for a year. When I told recruiters what I'd been doing I got "Oh.... Okay um... We'll call you."
I switched to rails for 6 months (same job) and recruiters were calling me non-stop. Working in low popularity langs limits my options. (even if they are great langs)
I'd say picking a language that's not yet popular limits your pool not just to the can-do's, but also to the can get my next job with this on my CV.
It's not object oriented, it's a functional language. It is polymorphic and all data is immutable. (Edit: actually realised its very much not like ruby, when I re-read this)
What I've seen in most projects is that the awesomeness that is threading (tasks and processes) is not really used. Supervisors aren't built as first class citizens.
I see teams eat the cost of NIH, but not get the benefits of crash fast, crash often.
So yes it does end up looking/smelling like ruby/rails but it really really shouldn't.
“A language that doesn't affect the way you think about programming, is not worth knowing.”
Took me a year before I found a dev/manager that even scratched at that power. A dev that pushed me to learn it. Before that I was a rails dev writing elixir. :)
I should mention there is no problem not going there. I respect my teammates. Ruby/Rails is a great way to get things done. I would 100% use it as my MVP lang.
I didn't want to confess, but yes you're right. Elixir (or the willingness to learn Elixir) is a great filter for me to find the right people. Again, this is my purely subjective opinion and might as well be true for other languages as well.
No one where I work has come in with Elixir experience. We look for people that are motivated to learn. Elixir has been one of the easier languages in my career to learn, and I think others that I work with would agree with that.
The degree of difficulty with learning Elixir/OTP depends on what one needs to learn. Phoenix web development has its own domain of learning requirements where as other domains have theirs. Learning Elixir is one thing but learning OTP, and then learning how to make sound decisions about design/architecture, is far more demanding of effort and time. "How do others solve this?" leading to github search tends to fall far short of what one would find with other language ecosystems.
Moving from Python/JS -> Elixir at a startup seems very high risk, have the rewards been worth it? I can imagine issues with hiring, unknown unknowns, and less libraries/support in general.
I know-- I was scared like hell. But after the first week and multiple staging releases in less than 10 days, I was on top of it. Yes, hiring is real challenge and am facing it already. However, if I, even as a startup, reach the salary threshold, then hiring is not a problem. I'd say I've yet not found a case where I couldn't find a library for my use case.
The start-up I work for uses Phoenix and Elixir. I've been there around two years now and we haven't experienced any issues with hiring or finding skilled people. If anything, I'd point to Elixir as being one of the major factors of our success.
We are a remote company, so things may not be as easy, depending on your location, if your company demands people be in the office.
Quite few I had (assuming OPs question to be broad enough)-
1. Reading Malcolm X's autobiography _The Autobiography of Malcolm X_ in college. It transformed me such that within that one week of reading it, I developed (given my low standards) the strongest sense of orderliness in my life. Additionally, I decided for myself 'to straighten myself up' for this life.
2. Reading Dostoevsky's _Brothers Karamazov_. My 'inner' transformation (at the age of around 20.5) was so immense that it was also apparent from the outside. My transformation was 'not to be surprised by bad/evil' and seeing good in everything.
3. In programming/ computer science/ functional programming/ mathematics (I still don't fully get LISP, Haskell et. al. like other people here) but there are encounters in the field of lambda calculus, computation, cryptography that have left me totally transformed. Too many to elaborate.
OP here. No matter how hard I try, I just can't get it until I make frequent visits from 'different sides' to a location. I didn't even know about Topographical Disorientation, but it really hits home. Anecdotally speaking, I always apologize to people around me saying that, "I think I don't have the location module in my brain".
My partner and parents know that I have this. I don't tell it to others.
I also go to the same location through many roads, but only once I have one etched into my mind.
I have to use G-Maps 7-8 times for visiting the same location if there are even days-wide gaps.
I am also bad at getting directions in maps. I have to orient and reorient one multiple times to figure out where I am, which way I am looking, amd which way I need to go.
I often just walk 10-20 meters to figure out which way I am looking or need to go.
Yeah, I am missing the "location module", too.
I have had this long before I had access to any sort of digital maps or GPS.
And, for me, Google Maps has been a lifesaver.
If I go with the help of it 8-10 times, I never ever forget the road again.
This is so weird for me given I have so good memory since my childhood. I could always memorize things real fast, without any superficial tools like mnemonics, flash cards, etc.
Thank-you, throwaway account and OP. It looks like "8" is a magic number--I tell others it takes me that many times of repeating to get around without GPS.
It's inconvenient for some things, like carpooling, and it can annoy close others to "never remember." Even if I try to focus on getting somewhere, attention never lasts for long.
I've tried heuristics and patterns, like "if I exited the highway and turned left, I need to turn right and merge to return home." But none seems to stick. I forget if I needed to turn left or right out of the retail place!
And this is a long way of saying I had to quit games like CounterStrike because of it, unable to really keep up with siblings and so on, so a part of my identity is wrapped up in this disorientation mode. I ask them how they "avoid getting lost," and to them it's second nature. They were already experts in it.
Maybe there is a better algorithm to orient oneself, something like when we need to print and "see all the env vars," to make it _so concrete_ (like Dr. Ng likes to say), that _now_ we can get to the heart of the problem we were working on.
I play only occasionally, and never the competitive sorts. I play CoD, Factorio, Age of Empires, etc. Played some Vice City, GTA-V back in the day.
In VC, GTA-V, I could never remember even the most common roads.
I always found myself looking at the map at the corner much more than the main screen when I needed to get anywhere.
If I played the competitive sorts, maybe I would discovered that I suck at those, too.
But when navigating IRL, I am a responsible driver, an alert "walker". Yet I never comprehend directions.
In the city, I always end up using the longer route than the shorter one, through more common and active places, because I know them more. I end up using the longer route most of the time. I have some "anchor points" in the city, and I always navigate through them, because I know those routes more. I don't risk other routes.
Today, I could not even recognize the route I just went through- when making the return trip (two hours stay). I had to ask around for directions. (Not my city)
Yet, besides this, my life is totally normal. Weird.
Author of the post here. Trying to respond to some of the comments. Firstly, apologies- I didn't intend to make anything click-baity. As a matter of fact, I didn't even post it here. I realized that I could have titled my post like "How to ask founders good questions".
But, and in all honesty, as a founder, it did/does feel such questions are existential. I used _search-engine_ as a mask, I'm actually building a deep-tech (I've commented/posted on HN before).
When you put everything on the line to build a startup for a mission you believe in to the extent that you can't even separate your existence from your startups 90% of the awake hours, then when such questions come from (this is important) "a specific set of stakeholders", it does become existential. I love being asked all these things from a random person or at a random pitch event. But when you're my colleague, co-founder, or my hero, or a potential investor who I've been in the loop with for weeks, then when you ask such questions, it does get hard.
Someone below suggested that such people probably shouldn't start a startup, but on the contrary, I wrote it not to whine about myself, but only with the sincere hope that if there's someone who is less tougher than I am, he/she would be more cared for (in the odd chance that someone reads it).
The true origin of this post was in my experience with building my indoor farming startup that will grow the best quality produce, cheaper, and way more resource efficiently than traditional agriculture. Ofc I've done my homework and I am actually taking a risk and doing the very uncomfortable by deciding to work on it. Imagine jumping back and forth between research-papers, circuit diagrams, dying-plants, finance, etc. etc., and then someone very relevant asks/tells you "you can't beat the oldest industry known to humankind", or "why can Infarm (a unicorn in the vertical) not do this?", "maybe just work on indoor grow lights (since it fits more into the standard lean startup model". In that moment, it seems existential. Maybe I interpret it wrong, but I interpret it as if you're telling me that my startup shouldn't (can't exist) and yes I'm so obsessed with my startup that I take it personally (again, only when it comes from specific stakeholders). I can give them the numbers, but if they say it can't be done or if it could have been done, it would have been done, then that's same as saying that I should shut up.
> But when your my colleague, co-founder, or my hero, or a potential investor who I've been in the loop with got weeks, then when you ask such questions, it does get hard.
I'm not a founder of anything, so I understand if this is just part of an experience that I've not had, but those are exactly the sorts of people who I would absolutely want to ask these sorts of questions. At least in my life, if I want to do something crazy, it's my close friends, my heroes, and my mentors that I go to first to get advice, and if I'm doing something that they don't see as sensible, I truly hope they tell me. I mean, they have done plenty of times already! And either I can give them a good answer and we can talk it over, or I can't give them a good answer, and then I know I need to think things over - either coming up with a better answer, or realising that what I was doing wasn't as sensible as I thought.
Likewise, I have friends where I hope I can be that person to them as well, and try and offer a critical perspective when I think they're viewing things in a rosier light than they realise. Obviously how you say that is always important, and in moments like that I try hard to frame my own perspective helpfully (and accept that my perspective may be missing something that they're seeing), but if there is a deep enough trust in a relationship, I think those sorts of things are possible and healthy.
I enjoyed your post and it resonated with me. Dumping cheap skepticism on a someone who is working hard to do something new is something I wish people would be more thoughtful about. It’s almost a reflex, but is it actually helpful?
The responses on this thread are kind of discouraging and missing nuance. It’s turned into “should you tell the hard truth or pretend the world is a fairy tale” (a false dichotomy) when it’s more like “don’t be a jerk in this moment”.
It's hard because some of the "easy questions" require a long answer - and sometimes you really ARE in a startup that shouldn't exist (or can't exist yet) - and your friends might be the people who need to pull you out.
But often, if it's someone who is "involved" or interested or even a co-founder, they really do have the question and want to know the answer - or at least be reassured that you've thought about it.
A startup is always in the strange place of trying to do something easy enough that a startup can do it, and hard enough that nobody has done it yet.
Waiting for someone to build something like this for founders of any promising startup out there (and not just YC startups). I totally understand the current choice of sticking to YC startups though. But do you imagine accepting non-YC startups sometime in the future?
I am a first-time founder of a deep-tech, and having secured pre-seed 6 months ago, and now that I have gathered all the evidence for the viability, I would definitely consider approaching something like 8vdX.
Sorry to read about such experiences people might have had, but I actually have had the exact opposite experience. I made few submissions/comments about 'dev' things, and because of the support of the HNers, I felt really good and was even more excited to work on them.
Second is my current moonshot I'm working on where I was expecting a lot of criticisms and negative comments, especially from programmers/devs in the AI space (Fully AI driven indoor farm, kind of deep mind for agriculture and understanding plants better than any human). But the largely positive feedback of the community was really energizing and uplifting.
I have posted this here before- hexafarms.com. I am trying to use ML to discover optimal phenotype for growing plants in vertical indoor farms to a. have the higest quality produce b. to lower the cost of producing leafy green/med plants, etc. within cities itself.
Basically, every leafy green (and herbs, and even mushrooms), can grow in a range of climatic condition (phenotype, roughly) ie temperature, humidity, water, CO2 level, pH, light (spectrum, duration and intensity) etc. As you might have seen around the world there is a rise in indoor vertical farms, but the truth is that 50% of those are not even profitable. My startup wants to discover the optimal parameters for each plant grown in our indoor vertical farm and eventually I would let our AI system control everything (something like alphaGo, but for growing plant X (lettuce, kale, chard, ). Think of it as reinforcement learning with live plants! I am betting on the fact that our startup will discover the 'plant recipes' and figure out the optimal parameters for the produce that we would grow. Then, the goal is that cities can grow food cheaper in more secure and sustainable way than our 'outsourced' approach in country side or far away lands.
So now I have secured some funding to be able to start working on optimizations, but I realized that *hardware* startups are such a different kind of beast (I am a good software product dev though, I think). Honestly, if anyone with experience in hardware related startups (or experience in the kind of venture I am in) would just want to meet me and advise me, I would take it any day. Being the star of the show, it's hard for me to handle market segmentation, tech dev, team, next round of funding, European tech landscape, etc. I am foreseeing so many ways that our decisions can kill my startup, all I need is advise from someone qualified/experienced enough. My email: david[at]hexafarms.com
Reminder to focus on nutritive content, flavour, and crop diversity, not just yield. The past 100 years of industrial scale agriculture, with the singular goal of maximizing yields, has done incredible harm. (This has come up on HN repeatedly, so I trust you've seen it, but it's worth championing)
I agree that micronutrient content has decreased in the past century. Some might be because of scale, some might be that yield gains are mostly driven by macronutrients and water, not micronutrients, it could be selecting varieties that taste better, or it could be depleting the soil.
That said, the US has an obesity epidemic, so there's no shortage of macronutrients. Macronutrient shortages also seem rare. Scurvy and rickets aren't exactly problems.
This isn’t an answer to your ML question, but it is an answer to your problem.
I heard about a greenhouse company that has programmed their climate control to match “best growing conditions historical weather”. So, they ask local experts what year / location had the best X and then they use that region’s historical weather and replay it in their greenhouse. I thought that was brilliant!
(Just realized this was Kimbal Musk that mentioned this)
When I studied farming back in 1998-1999 we once visited a greenhouse and one interesting thing I picked up was that by observation some gardeners had realized that lowering the temperature a bit extra an hour or two before sunrise they could get their flowers to be more compact instead of stretching.
This had replaced shortening hormones in modern gardening (or at least at that greenhouse, but my understanding they were just doing the same thing as everyone else).
I guess there is a lot more to learn for those who have scale enough to experiment and patience to follow through.
Sounds similar to what I read a long time ago about a big tomato farm in the Netherlands... Have you tried talking to actual farmers of that produce? Universities? Agricultural faculties do a lot of research in that direction.
Expensive, quickly perishable produce might be able to compete, otherwise I guess free water and energy from above in the "remote" classical farming will be hard to beat.
And then my naive guess would be that to generate enough data for a "ml" approach not only by name might be somewhat expensive.
This sounds so negative, but this is not my intention... I wish you all the best and hopefully will stumble upon a success story in the future :-)
I know this isn't going to sound as sexy as AlphaGo for plants, but I really think this is a classic multilinear optimization problem once you've properly labeled the data and defined the dynamics between the plants / other organisms (e.g., aquaponics). You're looking to optimize multiple variables across a set of known constraints and I think if you properly defined these constraints you could save a lot of headache / buildout by leveraging a pre-exsting toolset like Excel with the Excel Solver add-in an a couple hundred user defined functions. We're talking 1% of the work to get something useable and product-market-fitable with automatic output of graphs, etc, that clients could tune and play with locally without you needing to actually share the source sauce. Eventually you could switch to Python for something more dynamic / web based.
There's some great research on using evolutionary computation to explore plant growing recipes (light strength, how long to leave the lights on, etc). In one experiment, researchers discovered that basil doesn't need to sleep - it grows best with 24 hours of light per day. Risto Miikkulainen shared the experiment on Lex Fridman's podcast: https://youtu.be/CY_LEa9xQtg?t=27m7s I believe this is the paper describing that experiment: https://journals.plos.org/plosone/article?id=10.1371/journal...
This sort of ml problem is characterized by relatively expensive data labeling. Hence, hiring an expert or mixture of experts, and modeling the crop responses to their choices, will save you a lot of hill climbing The wrong part of the decision space
- CO2. Side note: I was surprised to find that most (all?) CO2 sensors used in closed plant production setups are not meant to operate below 400 ppm.
- Air temperature, pressure, relative humidity
- Photosynthetically Active Radiation (PAR)
- Addons like: wind speed, wind direction, soil moisture and Electrical Conductivity (EC)
- The coolest and most challenging: pH, EC, and flow rate
The hardest part has been running everything on battery while maintaining accuracy and using LTE (2–4G) and not common LPWAN options like LoRa. I'm primarily a software guy, so the learning curve has been huge.