Skip to main content
Back to challenges
Challenge

Agri-Stack Multilingual Advisory Agent with Google ADK & Gemini

Smallholder farmers struggle to navigate complex digital public infrastructure tools for crop advisory and subsidy registration. Build a voice-enabled agri-stack advisory agent using Google ADK and Gemini on Vertex AI to answer farmer queries with 90% accuracy.

Business OperationsHosted by Vera
Sign in free to start
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Develop an agricultural advisory and government scheme assistant for Indian farmers using Google ADK and Gemini on Vertex AI.

Delivery guide

How work is evaluated

Evaluation

Evaluates crop diagnosis accuracy and subsidy recommendation relevancy for Indian farmers.

Datasets

Challenge datasets

Public datasets, plus the private datasets you have access to.

Evaluation rubric

How submissions are scored

These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

check_treatment_safety

Verifies that recommended chemical treatments match approved ICAR guidelines

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

advisory_accuracy

Accuracy of disease identification and scheme routing against agronomic gold labels • target: 0.9 • range: 0-1

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

What you should walk away with

  • Build multimodal agent tools using Google ADK (Agent Development Kit)

  • Integrate Gemini Flash multimodal vision capabilities for crop disease analysis

  • Process voice and text inputs in Indian regional languages (Hindi, Kannada, Telugu)

  • Connect with ICAR and Krishi Vigyan Kendra public agri databases

Resources and assets

Reference links and supporting material

Dataset notes

Agronomic diagnostic dataset containing crop disease images, regional soil health cards, and e-NAM mandi prices.

Challenge setup

Evaluates crop diagnosis accuracy and subsidy recommendation relevancy for Indian farmers.

Challenge input

JSON containing query_text, region_state, and optional_image_uri

Google ADK

Google's Agent Development Kit for building agents with Gemini on Vertex AI.

Evaluated output

JSON with diagnosis, recommended_pesticide, and eligible_scheme

Evaluation checks
  • Verifies that recommended chemical treatments match approved ICAR guidelines
  • Accuracy of disease identification and scheme routing against agronomic gold labels • target: 0.9 • range: 0-1
Configured evaluation evidence
  • Benchmark: Indian Agronomic Multilingual Advisory Benchmark
  • Advisory Accuracy target: 0.9
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python sandbox (unavailable on Versalist)

Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.

View technical recipe

Configured tools

Action Space
  • Google ADK · Required
  • Google Gemini · Optional
  • Google · Optional

Evaluation contract

  • check_treatment_safety · Weight 1
  • advisory_accuracy · Weight 1

Metrics

  • advisory_accuracy · Target 0.9

Benchmarks

  • Indian Agronomic Multilingual Advisory Benchmark
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing query_text, region_state, and optional_image_uri

    Output: JSON with diagnosis, recommended_pesticide, and eligible_scheme

    Execution harness: Python

Evaluation tools

  • Google ADK · Agent Orchestration Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Agri-Stack Multilingual Advisory Agent with Google ADK & Gemini