- Climate · soil carbon
- Embedded hardware × Soil spectroscopy
- Closes 11 Dec
Hardware Engineer, Soil Sensing
Furrow Institute · Nairobi, hybrid · $60–80k
2,300today
2% of the way to the goal100kfarmers testing in the field by 2030
Where this role takes you
Furrow Institute's 2030 goal, and where the work is today
2,300farmers get a lab soil test each season
2% of the way to the goal100kfarmers testing in the field by 2030
Week 1
You read the null result.
Start with our 2023 Geoderma paper on why handheld spectroscopy failed. Then two days on the Rongai plots with Priya, scanning soil beside the cores going to the lab.
With
2023null result you start from
A soil test that costs $38 and takes three weeks will never reach every farmer, every season.
Every Furrow result rests on lab soil tests: 61,000 samples so far, each one dried, driven to Nakuru and analysed at about $38 a sample. Farmers wait three weeks for numbers that should change what they plant next week.
In 2023 we published a null result: cheap handheld spectroscopy was not accurate enough. This role tries again, properly. Design a $40 sensor, test it against the lab on real plots, and publish whether it works.
Is this you?
If two or more of these sound like you, you're worth talking to, whatever your CV says.
- You’ve built something that had to survive mud, heat and a dropped toolbox.
- You trust a calibration curve only after you’ve tried to break it.
- You’d rather publish a sensor that failed than oversell one that half works.
- You like explaining a circuit to a farmer as much as to an engineer.
Why this seat matters
“By the time the soil results came, I had already planted. Next season I want the number while I am still standing in the field.”
Peter Kiprono, farmer, Rongai
Meet the crew. There's a seat for you.
- You?This seatHardware Engineer, Soil Sensing
Priya Sharma · Sensors and field hardware, your manager
“I led the 2023 study that failed. I want you to prove me wrong with a better sensor.”
On a bad day: solders in silence with the door closed.
A Tuesday in Nairobi
Outside sampling season
- 09:00Bench test of the new optical board
- 10:30Calibration review with Marcus
- 12:30Lunch at the Kilimani workshop with the fabricators
- 14:00Firmware: a week of readings stored offline
- 16:00Call with Ayo’s field assistants in Rongai
- 17:30Lab demo: what broke this week
What life is like
- Fridays
- Demo at 4pm. Broken prototypes welcome.
- Every month
- Three days on the Rift Valley plots, transport paid.
- Each quarter
- A $1,500 parts budget you spend without asking.
- December
- Christmas week off for everyone.
What you'll do
- Design a soil carbon sensor that costs under $40
- Run side-by-side trials against lab tests on 200 plots
- Work with a Nairobi workshop to build it locally
- Publish the design and the accuracy data openly
What you bring
- 4+ years in embedded or sensor hardware
- Taken a design from prototype to a small batch
- Comfort with calibration and measurement error
- Happy to test in the field, in the mud
- No soil science background. Priya and the lab will teach you.
Your part in it
- DESIGN
- a soil carbon sensor that costs under $40 to build.
- TEST
- it beside the lab on 200 plots, against a bar Marcus sets first.
- WRITE
- firmware that stores a week of readings offline.
- BUILD
- small batches with the fabricators at the Kilimani workshop.
- PUBLISH
- the design and the accuracy data, whether it clears the bar or not.
What success looks like
- DAY 30
- you’ve read the 2023 null result and scanned soil beside the cores in Rongai.
- DAY 60
- Marcus has set the accuracy bar and your first board passes bench tests.
- DAY 90
- twenty prototypes are reading beside lab tests on 200 plots.
What you’ll need
- 4+ years in embedded or sensor hardware
- Taken a design from prototype to a small batch
- Comfort with calibration and measurement error
Nice to have, not required
- Spectroscopy or optical sensing. Priya will show you ours.
- Low-power firmware for devices that live outdoors
Where you’ll work
- OFFICE
- Nairobi, three days a week in the Kilimani workshop.
- REMOTE
- The other two days, from home in Kenya.
- FIELD
- Three days a month on the Rift Valley plots. Transport paid.
- HOURS
- East Africa Time. Core hours 10:00–16:00.
Languages
| Language | Need | Why |
|---|---|---|
| English | Required | The lab’s working language, and every design document. |
| Swahili | Nice to have | Ayo’s field assistants and the fabricators. Furrow pays for lessons. |
The essentials
Hardware Engineer, Soil Sensing · Furrow Institute
- Salary
- $60–80k
- Where
- Nairobi, 3 days in the workshop
- Leave
- 28 days + Christmas week
- Extras
- $1,500 parts budget a quarter
How you'll interview · about 2 weeks
- 1 · Chat with Priya
- 2 · Take apart our 2023 prototype (paid)
- 3 · Meet the lab
- 4 · A day on the Rongai plots
Your first stop
The Rongai plots, scanning beside the corers
Not ready to apply? Talk to Priya for 15 minutes.
No CV needed. Ask about the work, the people, or what a hard week looks like.