Deep-Tech · AI · Robotics · Agriculture

Decoding Plants.
Designing Resilience.

AI-powered plant intelligence for climate-resilient agriculture.

DeepPhenoTech combines plant science, AI, computer vision, robotics, and advanced sensing to transform crop data into actionable intelligence for breeding, crop monitoring, and precision crop protection.

Explore Our Technology
10+ yrsInternational plant science research
3 PlatformsRobotics, disease AI, cloud intelligence
Autonomous field robot with multispectral sensors operating between grapevine rows
Field platform in development — sensing, capture, and precision treatment
Wheat field

Climate change is reshaping the conditions crops have to survive in.

Most of what happens inside a field goes unmeasured — until it's too late to act.

DeepPhenoTech turns that gap into data — and data into decisions.

About DeepPhenoTech

Built on Real Biological Questions

DeepPhenoTech is a deep-tech company developing AI and robotic tools for plant breeding, precision farming, and cloud-based crop intelligence — bridging the gap between biological complexity and actionable agricultural decisions.

Vision

A future where every crop can be measured, understood, and managed with precision.

Mission

To advance climate-resilient agriculture through intelligent, precise, and scalable crop technologies.

Dr. Ajit Nehe
Founder

Dr. Ajit Nehe

Founder & Director

Plant Scientist & Agricultural Technology Innovator

A plant scientist and innovator with 10+ years of international research experience in plant science, crop physiology, plant breeding, and phenotyping. Holds a Ph.D. in Crop Physiology from the University of Nottingham, UK, with research experience across INRAE (France), CIMMYT (Turkey), and SLU (Sweden). His recent work combines plant science with AI, computer vision, robotics, and advanced phenotyping for climate-resilient crop production.

Sovit Rath

Sovit Rath

Lead AI/ML, Consultant

Supports AI model development, computer vision, machine-learning workflows, model training, and deployment.

Recognition & Research Support

Early Funding, and Collaborations That Support the Work

$26,000

First Prize

Pune Agri Innovation Hackathon, 2026

$11,000

Innovation Award

Sten K Johnson Foundation, Lund, Sweden, 2024

$32,000

Research Grant

Einar and Inga Nilsson's Foundation, Sweden, 2024 — for imaging system development and testing

Dr. Ajit Nehe holding the Pune Agri Hackathon winner's plate
First prize, Pune Agri Hackathon 2026
Award ceremony stage at the Pune Agri Hackathon
Ceremony, Pune Agri Hackathon 2026
Receiving the Sten K Johnson Foundation stipend certificate
Sten K Johnson Foundation Award, Lund, Sweden

Technology development and research support has also come through interactions with Sony R&D (Lund, Sweden), Wageningen University & Research, the University of Angers, the Swedish University of Agricultural Sciences, DCM Shriram Seeds, and Sahyadri Frame, Nashik. These reflect research support and technical interaction, not formal strategic partnerships unless stated.

Technology discussion with the Sony R&D team in Lund
With the Sony R&D team, Lund, Sweden
Sony and Lund University collaborative event on AI in agriculture
Sony & Lund University — AI in agriculture
What We Do

From Plants to Data to Decisions

Autonomous, AI-powered systems capture high-resolution crop information, analyse biological traits, and support more precise agricultural decisions.

01

Sense

Multisensor systems capture information from plants, soil, and the crop environment.

02

Capture

Robotics and computer vision collect high-resolution data at field, plot, canopy, or plant scale.

03

Analyse

AI and machine-learning models extract meaningful biological traits and detect patterns.

04

Understand

Growth dynamics over time, in relation to environment, become crop intelligence.

05

Act

Supports breeding, genomic prediction, crop modeling, monitoring, and precision agriculture.

The Problem

Agriculture Needs Better Crop Intelligence

Modern agriculture generates enormous biological and environmental variation, but much of it remains difficult to measure at scale.

Farmer checking crop data on a phone in the field

Limited real-time crop intelligence

Researchers and farmers often lack precise, timely information about crop health, growth, architecture, and yield potential.

Researcher manually recording crop measurements in a field

Manual crop assessment

Traditional crop monitoring and phenotyping can be slow, subjective, labour-intensive, and difficult to scale.

Diseased crop leaf, illustrating the case against blanket chemical spraying

Disease detection happens too late

Disease is often identified after visible symptoms spread, leading to unnecessary or blanket crop protection treatments.

Research field trial plots used for breeding trait data collection

Climate-resilient breeding needs better data

Breeders and seed companies need high-quality, high-throughput plant data to identify traits and accelerate crop improvement.

Our Technology

Built at the Intersection of Science and Engineering

DeepPhenoTech brings together biological science and advanced engineering to build intelligent systems for agriculture.

Plant Science

Plant physiology, crop breeding, plant architecture, crop health, agronomic and soil traits, and climate resilience.

Computer Vision

High-resolution imaging for plant detection, segmentation, trait extraction, and disease identification.

Artificial Intelligence

Machine learning and deep learning transform raw crop images and sensor data into biological information.

Robotics

Autonomous mobile platforms enable repeatable, scalable data collection in field or greenhouse.

Multisensor Sensing

RGB and multispectral cameras, LiDAR, RTK-GNSS, soil sensing, lodging-strength quantification, and light measurement.

Data Intelligence

Trait extraction, visualisation, stability analysis, selection support, and decision-support tools.

A Core Differentiator

Seeing the Crop From Inside

Conventional imaging focuses on the top of the canopy. DeepPhenoTech's approach is designed to collect information from within the canopy itself — CropScanalyzer can operate inside narrow crop rows, gathering in-canopy data from dense vegetation.

Field platform operating beneath a tabletop grapevine canopy, surrounded by hanging fruit
Operating beneath a tabletop-trellis canopy, in-row
Conventional View

Top-down imaging

Aerial and overhead sensors capture the upper canopy well, but miss what's happening inside dense vegetation.

DeepPhenoTech Approach

Inside-canopy data and intelligence

Designed to move through the row itself, capturing structures and micro-environments top-down imaging can't reach.

AI & Data

Turning Crop Images Into Biological Intelligence

Field → Sensors → Images → AI → Traits → Insights → Decisions. Our pipeline draws on computer vision, deep learning, image segmentation, object detection, and disease classification — including YOLO-based computer vision models and SAM-based segmentation workflows.

Products

Three Platforms, One Ecosystem

Sense → Analyse → Predict → Act. CropScanalyzer measures the crop, AIRadiBot protects it, and DeepAgCloud turns both data streams into decisions.

In development · Field validation

CropScanalyzer

AI-powered autonomous plant phenotyping platform

An autonomous phenotyping platform designed to collect high-resolution crop data and quantify plant traits for breeding, research, and crop modeling — including an outdoor field or indoor greenhouse robot. Designed to capture information from inside dense crop canopies, where conventional overhead imaging can miss important plant traits.

  • Autonomous field data collection, including inside-canopy acquisition
  • Plant architecture analysis and crop growth trait measurement
  • Light interception measurement and soil parameter sensing
  • High-throughput, multitrait phenotyping with data visualization

Target users: plant breeders, seed companies, agricultural universities, research institutes, crop science and agrochemical companies, and precision agriculture organizations.

Explore CropScanalyzer
Field phenotyping platform among grapevine rows
In development · Field validation

AIRadiBot

AI Disease Detection & Precision Treatment

Detect early. Map precisely. Treat selectively. AIRadiBot is designed to detect and map fungal disease before symptoms are visible to the eye, and to support selective, chemical-free radiation treatment rather than blanket spraying — reducing unnecessary chemical use, treatment cost, and environmental impact.

  • AI-based disease detection and early identification
  • Disease mapping and identification of affected field areas
  • Selective radiation-based treatment concept
  • Reduced dependence on blanket fungicide application
Explore AIRadiBot
AI-guided precision treatment system operating beneath a vineyard canopy
Platform under development

DeepAgCloud

AI-Powered Crop Data & Decision Intelligence Platform

DeepAgCloud is being developed as a cloud-based data intelligence platform for agricultural research, breeding, and precision crop management — bringing real-time visualisation, statistical analysis (via its Viewlysis module), and AI-based prediction into one place. Collect → Visualise → Analyse → Predict → Decide.

  • Real-time dashboards integrating CropScanalyzer, AIRadiBot, weather, and genomic data
  • Viewlysis: interactive statistical analysis for multi-trait selection and stability analysis
  • Prediction modules in development: yield, disease, genomic, quality, and best-parent
  • Breeding decision support: parent selection, hybrid development, trial analysis

Target users: plant breeding organisations, seed companies, agricultural universities, research institutes, and precision agriculture organisations.

Explore DeepAgCloud
Viewlysis statistical analysis dashboard, part of the DeepAgCloud platform
Applications

Built for Researchers, Breeders, and Precision Agriculture

Target applications include the following — many are part of ongoing field validation, not yet deployed across every crop.

Plant Breeding

Accelerate crop improvement with high-throughput phenotyping and quantitative plant data.

  • Trait measurement
  • Parent selection
  • Stability analysis
  • Phenotypic screening

Seed Companies

Generate objective crop data to support breeding, product development, and field evaluation.

Agricultural Research

Enable researchers to collect richer, more reproducible crop data at field scale.

Precision Agriculture

Use crop intelligence to support more targeted crop management.

Crop Health

Detect disease and crop stress earlier and support precision crop protection.

High-Value Horticulture

Target applications include grapes, tomatoes, strawberries, and vegetables.

Aerial view of replicated field trial plots used for breeding trait data collection
Replicated field trial plots — the kind of trait data breeding programs need at scale
Contact

Let's Build the Future of Crop Intelligence

We are interested in working with research institutions, universities, seed companies, agricultural organizations, technology companies, and partners developing the next generation of sustainable agriculture.

FounderDr. Ajit Nehe — Ph.D. Crop Physiology, University of Nottingham, UK

This opens your email app addressed to ajit@deepphenotech.com. See README.txt to connect it to a hosted form service instead.