Priyan Khandelwal
Product lead and integration
Apollo AgriVerse / GreenWorks Division / Smart India Hackathon 2026
Intelligent Soil
Decision Workbench.
A RAIT student team under D Y Patil University is building a transparent digital twin that compares root-zone water scenarios and explains the next irrigation action.
SIMULATED MVPHYDROGEL + BIOCHAR COMPARISONEXPLAINABLE RECOMMENDATIONSFIELD VALIDATION NEXT
The project question
When should a farmer irrigate - and why?
The team models the root zone using soil, crop, weather, irrigation history and optional sensor observations. Its purpose is one usable action: irrigate now, delay, apply a partial cycle, or inspect a reading.
01Do not treat AI as a black box.
Keep the model bound to a simple water balance and show the assumptions.
02Compare treatments honestly.
Hydrogel and biochar have distinct roles; neither is presented as a universal answer.
03Build trust through evidence.
All MVP outcomes are modelled until a controlled field comparison proves them.
RAIT SIH Team / D Y Patil University
A bounded student work package with clear ownership.
GreenWorks provides the industry direction and Apollo AgriVerse mentorship. Dr. Dipti Jadhav provides internal academic guidance at RAIT. The student team owns a separate SIH prototype, documentation and evidence-led handoff.
Divyanshi Bharuntia
Soil and material research
Zidane Madre
Water-balance state engine
Rudra Wagh
Crop and weather logic
Mubeen Sakharkar
UI and MVP demonstration
Siddhanth Muttoo
Confidence, QA and evidence
Working sequence
01
Frame
Freeze one crop, soil profiles, weather scenarios and the treatment comparison.
02
Model
Build a repeatable root-zone water balance with finite amendment behaviour.
03
Demonstrate
Show the scenario, trajectory, action, explanation and confidence in one flow.
04
Validate
Move to controlled plots only after the prototype is reproducible and reviewed.
SIH journey
One step cleared. The work continues.
Internal selection is recorded as a foundation for the next stage, with the demonstration and validation work kept visibly ahead.
- Step 1
Internal selection
The RAIT team presented Intelligent Soil and was selected to progress.
Cleared - Step 2
SIH-stage MVP
Prove the complete, explainable irrigation decision loop.
In preparation - Step 3
Evidence review
Test inputs, confidence states and the field-deployment path.
Planned - Step 4
Controlled validation
Move from modelled scenarios to documented field evidence.
Next horizon
The team represented its own work, Apollo AgriVerse and Chatake Innoworks with a clear project story. The record will grow only as each stage is completed.
Submission presentation
The SIH deck, ready to review.
Open the browser-native six-slide deck, download the editable presentation, or review the submission PDF.
Six-slide browser presentationOpen the complete SIH deck↗Student access
Everything the team needs in one place.
The full SIH working library is available here: submission content, team brief, roles, scope, script and source presentation.
Every Markdown document opens as a formatted reading view. Its original source remains one click away in the reader.
Open Intelligent Soil MVPScenario comparison and explainable recommendation→SIH workspace guideOverview of the working package→Six-slide submission contentSlide-by-slide SIH narrative→Team briefing and roadmapArchitecture, validation and economics→Roles and daily planClear work allocation for every member→Scope and mentoring modelWhat the external SIH team owns and what it does not→Team workflow and presentation scriptOpening, member scripts, final pitch and judge Q&A→
How this becomes credible.
Farmer trust is earned through a simple local explanation, manual control and controlled evidence - not through a label such as AI. The next stage compares control, hydrogel, biochar and combined treatments while recording irrigation, soil moisture, crop response, cost and usability.