Insectarium X

We Built a Better Home for the Bugs That Eat Pests

Why a software company spent 18 months on a climate chamber for beneficial insects, and where it could go next.

Some of the best pest control in farming has six legs.

Tiny wasps that lay their eggs in whitefly larvae. Predatory mites that eat spider mites for breakfast, lunch and dinner. Lacewing larvae that go through aphids like popcorn. Growers call them bioagents, and every one of them is a reason to spray a little less.

There's a catch, though. Before they can protect a crop, someone has to raise them. And they are very fussy houseguests.

The good bugs need the right weather

Raising beneficial insects means holding a small room at exactly the right temperature, humidity, light and airflow. Then changing all of that as the insects grow, and again when day turns into night.

Today a lot of that is done by hand. Someone checks a thermometer, nudges a heater, writes a number in a notebook, and hopes nothing drifts over the weekend. It's slow work, it's hard to prove afterwards, and one bad night can ruin a whole rearing cycle.

When rearing is unpredictable, biological control stays a niche. When it's predictable, more farms can try it.

That was the problem we wanted to take off people's hands.

So we built a room that runs itself

Insectarium X is a closed chamber for raising bioagents, plus the software that runs it. You give it a rearing program, split into phases that follow the insect's life stages, each with its own day and night settings. It does the rest: heating, cooling, humidity, ventilation and lighting, all measured and adjusted on its own.

You can drive it from the touchscreen on top, or from a laptop or phone through a web app. Every run and every change is written down, so months later you can still answer the question every quality manager asks: under what conditions was this batch raised?

Two decisions shaped most of the design, and both came from asking “what happens when things go wrong?”

The internet drops. The chamber is built in three layers: the device itself, the software that runs on it, and the cloud. The part that controls heaters and fans doesn't need the other two. If the connection goes down, the insects never notice. The device keeps the program running, stores its readings, and syncs them once it's back online.

The power goes out. Most simple timers start again from zero. Ours works out how much time has actually passed and picks up in the phase it should be in by now. A three-hour cut on day nine doesn't send your insects back to day one.

The insects don't care about your Wi-Fi. So we made sure the chamber doesn't either.

What 18 months and one honest outsider taught us

We worked on Insectarium X from March 2025 to September 2026. The first stretch was lab work: testing controllers, sensors and heating parts until we trusted them, and designing the chamber around them. Then we built the prototype, ran it hard in the lab, and handed it to someone who had never seen it before.

That last step mattered most. An outside quality-management expert used the system on their own and told us what was clumsy. Their verdict was that it's ready for lab work and for teaching students. Their notes went straight into the next round of fixes.

A few numbers from the testing:

• 0.4 °CThe largest gap between our chamber and an independent reference instrument. The target was within 1.0 °C.
• 1.5 pointsThe largest humidity gap against the same instrument. The target was within 5.
• ~30 secondsTo reconnect to the cloud after a five-minute network outage. The program never paused.
• TRL 7The prototype has been shown working in a real operating environment, not just on our bench.

It isn't finished, and we'd rather say so. Automatic humidity control still needs work. So do full syncing of data recorded while offline, exportable history reports, and a wider cooling range.

Where this could go next

A chamber that holds the weather steady and remembers everything is useful in more places than we first planned for. Here's where we see it fitting, all inside the same world of growing things without chemicals:

Rearing on the farm, not just buying in. A greenhouse grower could keep a small, steady supply of a local beneficial ready to release the moment a pest shows up, instead of waiting for a delivery.

A paper trail for every batch. For companies that produce bioagents, each batch could ship with its own climate record. That's a quality certificate written by the machine, not filled in afterwards.

Recipes worth sharing. Rearing programs can already be shared inside an organization. A tested program for a particular wasp or mite could travel from a research lab to a dozen farms, so nobody has to start from scratch.

Side-by-side experiments. A research lab could run the same species in several chambers at slightly different humidity and compare results that are fully logged. No more arguing over whose notebook to trust.

A teaching lab that runs itself overnight. Universities could let students design a program, watch it from their phones, and read the full record the next morning. Our outside tester already pointed in this direction.

Pest control companies with their own test colonies. Firms that work in pest control could keep reference insect populations in stable conditions to test their treatments.

Every one of these ideas starts with the same thing: conditions you can trust and a record you can show.

Who we'd like to meet

We're looking for farmers, labs and universities who want to try Insectarium X or help us shape what it becomes. If you raise beneficial insects, teach people how to, or just want fewer chemicals on your crops, we'd like to hear what you'd need from a chamber like this.

Write to us

The fussiest houseguests in farming deserve a proper home. We think we've built a good first one.