Three Weeks, Two Sensors, One Lesson

Last summer, we built working gold and copper sensors in about three weeks.

That sentence is short, but it sits at a turning point for the company, so it's worth unpacking.

For most of the prior year, our work had centered on arsenic. There were good reasons for that. Arsenic is a clear environmental contaminant and a real problem at military sites, in groundwater, and around legacy industrial activity. The biology of arsenic-responsive sensors is well-understood, so that's where we started.

When we began talking seriously with mining teams, the response was consistent. Arsenic was interesting, but it was not what they needed to measure. They needed gold, copper, and the other commodities that drive their economics. Arsenic mattered as a contaminant, but it would not fund a drilling program or change a milling decision. They wanted to measure the things that pay for the work.

Fair enough.

So we redirected the platform toward gold and copper, and we could move quickly because we were not starting from zero. The three weeks rested on more than a year of prior work: a working sensor architecture, a validated hyperspectral reporter system, and a library of microbial chassis we had already characterized. The new work was the part we did not yet have: the biological machinery to detect gold and copper directly, and the engineering to make those readouts robust enough to be useful.

That's what the three weeks were for.

The short technical version: every biosensor has two parts. A sensing module that recognizes the target chemistry, and a reporting module that produces a detectable signal when the sensor is triggered. We could keep the reporting side largely intact and focus the engineering on the sensing side, specifically on metal-responsive regulatory systems that would respond cleanly to gold and copper at relevant concentrations.

"Cleanly" is doing a lot of work in that sentence. Mining environments are messy. Other metals are present. Soil chemistry interferes. A sensor that lights up for gold and for everything that looks vaguely like gold is not a sensor anyone can use. So the engineering focus was on specificity, on signal strength, and on getting a response that was readable in realistic backgrounds and not just in pristine lab water.

It worked. Not perfectly from the start, but well enough to demonstrate that we could engineer toward the commodities mining teams actually care about, on a timeline that matches how those teams operate.

The broader point is more interesting than the sensors themselves.

The lesson wasn't "we can engineer biosensors fast." We already knew that. The lesson was that the question isn't what's possible; it's what's worth building. And the answer to that question doesn't come from the lab. It comes from the people who would actually use what you're making.

Three weeks didn't change our technology. They changed what we pointed it at.

That's why this matters. Not as a technical milestone, but as a focus shift from "what we can detect" to "what's worth detecting."

—Brandon Fields, PhD, Chief Science Officer