31,163 auto-generated species pages
Written, in the sitemap, and deleted. They'd have ranked for a while and given away the catalogue — which is the product's asset, not the site's content.
Two people, one studio
Built by a small studio that decided to do the boring parts properly
LeafRanger is made by Country Devs LLP. There is no growth team, no advisory board and no round to announce — there are two people, a long list of decisions about what the app is not allowed to do, and a catalogue we refused to fake our way through.
LR
Pressed once, and it doesn’t come out.
What we’re building
Five parts, and they only work bolted together
Any one of these exists elsewhere. The reason to build all five is that a plant assistant without a real catalogue guesses, and a catalogue without a home map can’t tell you anything about your windowsill.
A graded catalogue
381,000 species, each marked by how well it's actually known — and researched on demand when someone adds one we only hold a name for.
An assistant that reads it
Herb answers from your plants' records rather than from a model's memory, and says so when the record is empty.
Your space, modelled
Homes, rooms, spots and light quality, with watering that moves with your actual weather.
People who grow things
A community for the questions a database can't answer, moderated rather than left to rot.
Somewhere to buy them
Independent shops whose listings bind to the same catalogue record the app cares from.
What holds it up
Four rules the product is built on
Not values on a wall. Each of these is enforced somewhere in the code, and each one has cost us a feature or a nicer-sounding sentence at some point.
Say what we don't know
An empty field renders as empty. Herb refuses to state a number it doesn't hold, and identification declines rather than guessing.
Research before refusing
A plant we hold only a name for gets looked up the moment someone adds it — about five seconds. Only a genuine dead end produces "we don't know".
Never let a model touch safety
Toxicity is written by a checked source or a person. No AI, at any tier, even when the field is empty.
Don't sell what isn't built
Several features listed in our own tier config are absent from this site because nothing implements them yet.
The discard pile
Things we built and then threw away
A page about principles is cheap. Here is what they actually cost — five decisions that each undid work that was already done.
Citing our data sources
Every source name came off the site. Provenance reads like a trust badge and works like a recipe for a competitor.
A “pet-safe plants” listicle
The easiest traffic on this whole site, and it would require treating “nobody has assessed this” as “safe”. We won't.
Tier features we'd already written down
Seasonal forecasting, priority support, full assistant history — all in our own config, none implemented. They're absent from the pricing page.
A launch date
Repeatedly. We don't have one we'd bet on, so the waitlist says so instead of picking a month that sounds close.
Where it honestly is
Nothing has launched. Nobody has paid us anything.
The rest of this site is written with some confidence, so it’s worth being blunt here: there are no users, no sellers and no revenue. What exists is a built app that isn’t released yet.
you are here
Built and working
- The catalogue, graded and researching its own gaps
- Identification, diagnosis and weather-aware watering
- Homes, rooms, spots and light
- Herb, rebuilt to answer only from what we hold
- Community, and the marketplace mechanics
Not done
- Billing switched on
- Store listing and release
- Vendor applications opened
- A data export that actually produces a file
Who builds it
Two people, and you can look both of us up
Not a collective, not a stock photo of a team that doesn’t exist. If something in this app is wrong, one of these two got it wrong.
NameMayur Srivatsav
RoleChief Executive Officer
Practising8+ years
- Product strategy
- Systems thinking
- UX architecture
Decides what LeafRanger is for, and what it stays out of. The product's refusals — no ads, no invented facts, no feature we can't stand behind — are strategy decisions before they're engineering ones.
Check the claim on LinkedIn
NameSurya Kumar
RoleChief Technology Officer
Practising10+ years
- Technical architecture
- Distributed systems
- Full-stack engineering
Builds it. The graded catalogue, the on-demand species research, the rule that no model may ever write a toxicity record — those are architecture, and they're the reason the app can say “I don't know” instead of guessing.
Check the claim on LinkedInHow it gets paid for
Two ways in, and three we’ve ruled out
Money in
Paid subscriptionsTwo monthly tiers, for people who want more room and conversations with Herb. The free tier is funded by them.
Marketplace commissionA published percentage on physical goods sold through the shop. Not decided yet, and it'll be in the vendor terms before anyone lists anything.
Ruled out
AdvertisingThere is no ad surface anywhere in the app and we're not building one.
Selling what we learn about youYour garden, photographs and conversations aren't a product, and aren't handed over for training.
Charging for the basicsIdentification, watering, the whole library and pet safety stay on the free tier permanently.
Nobody has paid us anything yet, so none of this is revenue — it’s the shape of the business we intend to run, written down before there’s money involved to argue with it.
Fair questions
Including the awkward ones
- Who actually makes LeafRanger?
- Country Devs LLP — a small studio. Two people work on it: Mayur Srivatsav, who runs the company and decides what the product is for, and Surya Kumar, who builds it. Both have public LinkedIn profiles linked above, so you can check rather than take our word for it.
- Two people? Is that enough to trust with my data?
- It's a fair question, and the honest answer is that size cuts both ways — there's no separate growth team quietly adding an ad network, and there's also nobody to take over if we stop. What we can point at is what the product refuses to do, and the fact that those refusals are enforced in code rather than promised on a page.
- Is this a side project?
- It's a product from a working studio, built to be released and maintained rather than shipped and abandoned. We're not going to dress that up with a funding announcement we haven't had.
- How many users do you have?
- None. Nothing has launched, nobody has been charged, and there are no sellers. Any number you see on a pre-launch site is either a waitlist count dressed up as traction or an outright invention.
- How do you make money?
- Two paid subscription tiers, and a commission on physical goods sold through the marketplace. Not advertising, and not by selling what we learn about your garden — both are ruled out rather than merely unplanned.
- What happens if you stop working on it?
- We'd rather say this plainly than have you find out: a two-person studio is a real risk, and no reassuring sentence changes that. What we can control is not building a product that traps what you put into it — a data export is one of the things that has to be finished before we'd call this released, and it isn't yet.
- Why should I believe the honesty claims on this site?
- Don't take them on faith — check them. Ask Herb about a plant we hold no data for and watch it decline to invent one. The refusals are the part you can test, which is the only reason they're worth writing down.
- Where are you based, and how do I get in touch?
- Country Devs LLP, working worldwide. Full registered details and contact routes go on the legal pages, and we'd rather leave those blank until they're accurate than fill them with a placeholder.
The door’s open, the shop isn’t yet
LeafRanger isn’t released. If what you’ve read here sounds like an app you’d want, leave an address and we’ll write once — on the day it’s actually installable.