PULSE · v17.3
Applied irrigation
A trained regression model predicts how much irrigation was applied.
Detailed evaluation available in private diligence materials.
For investors · Series Seed · September 5, 2026
The labeled corpus that turns permanent-crop sensors into a defensible water decision every week.
Irrigation intelligence for California permanent crops, with trained models on our own data and a certified crop adviser in the field every week.
2 paying customers. $320K signed ARR on about 4,000 contracted acres. 8,000 verbal acres contingent on funded sensor installation.
Use the sensors and crew you already have with BYOS, after we confirm the sensor connection. AgWaterAI adds the intelligence and the record. With Managed, we install and own the sensors and run the field program.
Your adviser reviews the evidence and makes the irrigation call. The signed record shows what was recorded and whether it changed after signing; it does not prove a measurement was correct.
The problem
Under SGMA, groundwater agencies allocate water and charge for overuse. A 1,000-acre pistachio operation running 1.5 acre-feet per acre over its allocation in East Kaweah GSA faces $750,000 in a single year, before state fees.
Illustrative over-allocation example
$750,000
1,500 acre-feet × $500 per acre-foot. East Kaweah GSA's board-adopted AY2026 penalty tier, verified August 2026 in the deck sources.
What we built
The adviser assigns the field work. Crew captures chamber readings and decisions in the app. The adviser reviews the exceptions and writes the order in the units the irrigator uses: pump hours, inches on the block and acre-feet against allocation.
Stress outputs need local chamber readings. No trained model's score enters the order. The adviser turns the field evidence into the irrigation decision.
Follow the field-team walkthroughPULSE · v17.3
A trained regression model predicts how much irrigation was applied.
Detailed evaluation available in private diligence materials.
SENTRY · v52
A trained classification model helps the adviser choose which blocks to check. Local chamber history is required; the model covers the days between visits.
Detailed evaluation available in private diligence materials.
WATCHDOG · v20
A trained classification model flags anomalous sensor behavior for review before it reaches a decision.
Detailed evaluation available in private diligence materials.
Detailed results and their limits are available in private diligence materials; they are not a promise of performance on a new farm. PRESCRIBE ships mined rules with trained successors in shadow; a physics planning alpha projects days to the depletion line; SALT FLUSH is trained but gated. No published number is claimed for these.
The water-order record is signed when the adviser approves the order. The signature proves who issued a record and whether it changed since; it does not replace meter or sensor validation. Patent-pending, assignments in process.
We install and own the sensors and run the field program. Managed generates the paired field labels that build the intelligence.
The grower connects sensors already owned. BYOS distributes that intelligence across more acreage. Field visits and pressure-chamber work are billed separately.
Proof · September 2026
2 paying customers
$320K signed ARR
About 4,000 contracted acres
Both customers are on Managed at $80 per acre per year. One contract is written as a 2026 season fee and is counted as ARR because the service renews annually.
68 active in the September 2026 snapshot: reporting within 48 hours in the August 30, 2026 logger pull. The network includes long-term training sites beyond the paying customers.
6,097 dated pressure-chamber readings collected; 3,844 in the joined stress matrix.
Daily multi-depth soil data since 2016 on the oldest site, with about 240 site-years on the live fleet.
Strategic direction
This is a direction, not a market-entry commitment or a claim of district, buyer or registry acceptance; no credits have been issued.
The measured moat · September 3 to 4, 2026
We compared a farm-only stress model with our corpus model on the estate.
The corpus gives a new farm a head start on stress calls, and the farm’s own chamber history takes over.
Research result, not a customer promise. The research and its qualifiers are available in the investor data room.
Applied-water prediction is not a moat: the competitor matches it after one season. No water or dollar savings are attributed to these models.
The pipeline
Verbal 3-year terms at $80 per acre, subject to execution. Citrus, almond and pistachio across Kern, Tulare, Kings, Fresno and Madera counties.
Post-deployment scenario
+$640K on conversion. About $960K annual run rate on about 12,000 acres.
The additional revenue depends on funded installation and conversion of the verbal commitments.
The team
Advisor · Certified Crop Adviser
Founded the family irrigation consulting practice in 1984. 40+ years in the Central Valley.
CEO · Certified Crop Adviser #384333
16 years managing irrigation across the Central Valley. 22,000+ acres of irrigation consulting. His chamber program produces the labels; both customers came through his relationships.
CTO · AI engineer
Built the models, pipelines, dashboards and iPhone app.
Use of funds · 18-month plan
The 8,000-acre pipeline, contingent on installation.
18-month total.
October 2026 to September 2027. University participation is not committed.
First conversion bar: February 17, 2027.
6-month bridge.
Total: $1.17M
Planning assumption: about 90% grant recovery on sensor deployment, not a commitment. Awards, coverage and timing vary and are never guaranteed. Grant assistance and subscription are separate services, never tied.
The ask
$11.7M post-money · 10% dilution · Delaware C-Corp
Book an investor diligence callOpen the data room · One form, no NDA.
Not an offer to sell securities. Forward-looking statements are not guarantees. No water or dollar savings are attributed to the models; no crop ROI is claimed; no water credit has been issued.