Who we are
We were founded by an engineer and a third-generation cattle rancher. Richard J Albertini, Co-Founder & COO, brings twenty-five years of FDA- and ISO 13485-regulated operations from Johnson & Johnson, ThermoFisher and GE Medical, where he is a GE Six Sigma Master Black Belt and a student of the Toyota Production System. The discipline he carries out of Johnson & Johnson is not only Six Sigma and Lean — it is the beyond-compliance culture of the J&J Credo and the 1982 Tylenol response, where the company recalled 31 million bottles at roughly $100 million rather than do the legal minimum, on the principle that a company’s first responsibility is to the people who use its products. We engineer every regulated operation to that standard, and we bring it now to the small-and-medium farm.
Rafael Diaz-Albertini, Co-Founder & CEO, is the creator of both platforms and the proof they already run on a real farm. A biologist by training but a farmer, operator and trainer by trade — a third-generation cattle rancher, Lean Six Sigma Black Belt and certified SAP Professional, with formal training in agroforestry and in dairy production and management. He owns and runs Industrias Agripalma, a 400-hectare organic, regenerative palm-oil and cattle operation in Mérida, Venezuela, where every AgroPulseAI feature is dogfooded daily, in Spanish, against real livestock, real regenerative crops and real veterinary-and-regulatory complexity before any external user sees it.
Both founders are Venezuelan-born of the same family, Spanish-first and fluent in English — which is what grounds the platform’s Latin America reach in lived experience rather than market research. The shared operational-discipline vocabulary — value-stream mapping, root-cause analysis, statistical process control, kaizen, SAP — applied by Rafael in real farming rather than a back office, is why this engineer-meets-farmer partnership doesn’t carry the translation bottleneck that defeats most such pairings.
The market we saw
Eighty-eight per cent of US farms are small-scale, with single-digit adoption of precision agriculture, while large farms operate at 70%-plus. The UK Cattle Tracing System, introduced in 1998, is in the UK Government’s own words “no longer fit for purpose,” still based on information and documents sent by post and processed manually. The UK’s own ambition is for its SMEs to become the most digitally capable in the G7 by 2035 — and the phrasing alone implies the gap. Within UK SMEs, agriculture is one of the most behind.
While modern industries were transformed by CRM, ERP, Lean Six Sigma, project management and real-time dashboards, small-and-medium farming still runs on years of experience, tribal knowledge, paper records and manual data entry — roughly thirty to forty years behind where modern industrial process technology sits. The friction is everywhere: soil testing, spray decisions, mastitis records on 150 cows that live in notebooks, vet visits priced at £150–300 that arrive after the herd is already sick, regulatory paperwork that consumes weeks per year per farm, and data that exists everywhere but cannot be analysed because no one has the time. That gap is the structural opportunity.
The platform: engineered by farmer ask
AgroPulseAI didn’t start as a product. It started as Rafael’s own workaround. In late 2023, running the family farm in Venezuela from thousands of miles away while in a Lean Six Sigma role in the UK, he could never get the numbers in front of him when a decision had to be made — so he built a platform to do it himself. As Harvest Manager at Agrial Fresh Farms in 2024 he was doing five hours of manual data entry a night. In pig production at Rattlerow Farms in 2025 he was writing data on the window of the buggy to stay Red Tractor audit-ready. He built tool after tool, moved the whole thing onto WhatsApp — where the farmer already lives — and in January 2026 left to build AgroInsights full-time.
The AI is the platform. AgroPulseAI takes input from any device — a sensor, a phone, a photo, a voice note, an EID reader, a regulator’s web form, a buyer’s PDF — and its agentic models distil that data into the next decision the farmer or vet needs to make. Critically, the data flows automatically. No farmer typing into an iPad. Sensor signal, EID reads, WhatsApp photos, weather feeds, soil probes: all of it autonomous. The agentic AI watches, anticipates, decides and surfaces; the farmer acts on what the platform brings them.
Every product since has been a response to a farmer ask. AgroHealthAI was born when vets asked how to extend coverage with fewer in-person visits — an AI veterinary clinical decision-support engine that surfaces a ranked differential across thirteen species in under two seconds, grounded in a curated knowledge base of more than 3,000 peer-reviewed sources, Bayesian-calibrated and ensemble-validated. The vet sees the evidence chain, not just the answer, and the vet makes the diagnosis. AgroHealthAI is extending into ISF biosensors — interstitial-fluid sensors that monitor animal health continuously, in the animal — to give the vet lead time before sub-clinical signals present. These are patent-pending and not yet commercially available; sensor detachment under 3% over a wear cycle is a simulation target, and a bovine ISF correlation sub-study is the work built to establish the real lead-times. AgroControlsAI carries our on-farm hardware, including the Pulse Edge gateway — born when farmers asked about free-range farms with no Wi-Fi — which aggregates and syncs when bandwidth allows. Pulse Winery, within AgroPulseAI, is a UK-native vineyard-to-bottle platform that applies Six Sigma to vintage variation. AgroGenesAI closes the loop between today’s sensor data and next year’s bloodlines, and AINS Companion extends the approach to non-food animals.
A flywheel that turns in both directions of time
The platform’s network effect compounds the way a connected fleet does. Every farm that joins feeds the central AI more signal, every minute, from every species, in every condition — a mastitis signature picked up on a Suffolk dairy farm at 5am improves the early-detection algorithm for a Wisconsin dairy farm by 9am the same day.
But we also do something a forward-only platform cannot: on onboarding we systematically ingest the farm’s existing records — OCR for the hand-written, structured parsers for the digital, voice-and-photo intake for the records the farmer prefers to narrate — into the farm’s local knowledge layer. A dairy farm with thirty years of milk records, a vineyard with twenty years of harvest logs, a vet practice with decades of clinical history: all of it becomes queryable from week one. The farmer doesn’t wait five years for the platform to know their farm. It knows in days.
And there is a third flywheel, the one that may matter most, because it isn’t about any single farm’s data at all. Each specialist agent continuously reads the world’s published science in its own field — the peer-reviewed journals, the veterinary, agronomic and genetics literature — and is engineered to keep becoming an expert who has read everything in its field and forgotten none of it. When the next anomaly arrives, the expert work is already done: the farmer or vet receives an evidence-anchored answer with its sources attached, and that reasoning is itself written to the verified-provenance rail. A competitor can buy sensors and write software; what it cannot quickly stand up is a faculty of domain-expert agents that have already read the world’s literature and compounded that reading for years. This is a knowledge moat sitting on top of a data moat.
The verified-provenance rail — and Outcome Paper™
Beneath every pillar runs the cryptographic verified-provenance rail. Every meaningful event is hashed, aggregated hourly, signed and anchored to public blockchain — with Hedera as our primary network. Verification is by inclusion proof: a regulator, a buyer, an insurer or a consumer can verify any event without ever needing access to the farmer’s raw data. The regulator gets the proof, and the farmer keeps the data sovereign.
This matters because the world wants proof. A beef farmer can no longer simply say “trust me, the withdrawal period was respected.” A vineyard cannot just say “trust me, this isn’t a fake” — the global wine counterfeit market runs to over $3 billion a year. A vet can no longer just assert that antimicrobial stewardship was followed. Verified-provenance solves all of this at the moment the event happens, at the source where the data is born.
On top of that rail we’re developing Outcome Paper™ — an outcome-based, performance-fee commercial framework that lets small-and-medium farmers capture fair value for verified outcomes: the kind of financial infrastructure the 1848 Chicago Board of Trade gave only to large operators. Forward-priced verified-welfare beef. Provenance-verified harvest contracts. Environmental-stewardship-verified pasture management. The rail is in production; the commercial framework is in development.
Built beyond compliance — into the regulatory tailwind
The regulatory environment is moving in our direction — named, dated and funded. The UK 5-Year AMR National Action Plan 2024–2029 needs auditable per-animal data for targeted antimicrobial use. The EU’s Farm-to-Fork strategy requires verifiable supply chains for its 50% antibiotic-reduction target by 2030. The UK 25-Year Environment Plan requires environmental-stewardship evidence. The UK Pesticides National Action Plan 2025 commits to a 10% reduction in farm pesticide use by 2030 via precision technology — exactly what photo-based diagnosis plus precision spraying delivers. And the UKIPO Green Channel accepted our two oldest patent applications on environmental-benefit grounds.
Our operating posture meets these standards before they require us to. We run a documented Quality Management System voluntarily aligned to ISO 9001, ISO 13485, ISO 27001 and IEC 62304 — the disciplines that govern medical devices — none of which is required by current UK or EU veterinary regulation. We built it because it is the right way to engineer a platform that vets, farmers, regulators and insurers will increasingly rely on. When the regulatory expectations catch up to where we already operate — and they are catching up fast — we won’t be retrofitting. We’ll already be there. That is the Johnson & Johnson Credo legacy, transposed into agritech.
Where we go
We hold eleven inventive patent families, with nine applications on file — six in the UK and three in the US — and our two oldest UK applications accelerated through the UKIPO Green Channel. We never describe a pending application as granted or protected.
The vision: every small-and-medium farm, in every geography we can reach, with a digital twin in its pocket — the engineer’s discipline and the farmer’s lived knowledge available to every operator who could never before afford either. We will be the brain: not the OEM, not the sensor manufacturer, not the silicon foundry, but the agentic AI that decides the next action, the verified-provenance rail that anchors every event, and the federated learning that gets smarter with every farm onboarded.
Farming is roughly thirty to forty years behind the rest of the modern economy. We are going to close that gap — one farmer ask at a time, one product at a time, one cryptographic proof at a time. Engineer and farmer, side by side, because neither of us could do it alone.
Predict. Prevent. Grow.
— Rafael Diaz-Albertini, Co-Founder & CEO · Richard J Albertini, Co-Founder & COO
AgroInsights Technologies Ltd · Stowmarket, Suffolk, United Kingdom