Skip to main content
Loading navigation
Back to challenges

Build an Agritech Crop Yield & Livestock Diagnostics Agent with LlamaIndex

Agritech startups like Verdant Impact are modernizing rural agricultural productivity in India. In this challenge, you will implement an agritech soil health and livestock disease diagnostic assistant using LlamaIndex RAG pipelines. The system indexes agricultural extension research, veterinary manuals, and local soil moisture telemetry to deliver actionable crop rotation and livestock health interventions directly to rural farmers in regional languages.

Business OperationsHosted by Vera
Sign in free to start

What you are building

Build an agritech crop yield and livestock diagnostic RAG pipeline using LlamaIndex.

How work is evaluated

Evaluation

Evaluates accuracy of agronomic recommendations based on soil NPK inputs.

Datasets

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

How submissions are scored

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

ph_correction_test

Ensures alkaline soil (pH > 8.0) triggers Gypsum recommendation.

Binary check

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

Dimension 2

agronomic_accuracy

Correctness of fertilizer dosage recommendation. • target: 0.95 • 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

  • Index agricultural and veterinary diagnostic documents into LlamaIndex vector stores.

  • Implement multi-document retrieval with hybrid keyword-vector query routing.

  • Evaluate soil NPK values and generate custom fertilizer dosage schedules.

  • Deliver localized agronomy recommendations.

Resources

Dataset notes

Agronomic trial reports and soil nutrient requirement guidelines for Indian crop varieties.

Challenge setup

Evaluates accuracy of agronomic recommendations based on soil NPK inputs.

Challenge input

JSON with crop_type, nitrogen_ppm, phosphorus_ppm, potassium_ppm, soil_ph.

LlamaIndex

Required data framework for agronomy retrieval.

Evaluated output

JSON with recommended_urea_kg_per_acre, recommended_dap_kg_per_acre, ph_correction.

Evaluation checks
  • Ensures alkaline soil (pH > 8.0) triggers Gypsum recommendation.
  • Correctness of fertilizer dosage recommendation. • target: 0.95 • range: 0-1
Configured evaluation evidence
  • Benchmark: Agritech Precision Benchmark
  • Agronomic Accuracy target: 0.95
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python sandbox (unavailable on Versalist)

Shows the first task, up to three tools, and up to three evidence items.

View technical recipe

Configured tools

Action Space
  • Llama Index · Required
  • Agno · Optional
  • Cekura · Optional
Orchestration
  • Agno · Optional

Evaluation contract

  • ph_correction_test · Weight 1
  • agronomic_accuracy · Weight 1

Metrics

  • agronomic_accuracy · Target 0.95

Benchmarks

  • Agritech Precision Benchmark
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON with crop_type, nitrogen_ppm, phosphorus_ppm, potassium_ppm, soil_ph.

    Output: JSON with recommended_urea_kg_per_acre, recommended_dap_kg_per_acre, ph_correction.

    Execution harness: Python

Evaluation tools

  • LlamaIndex · RAG Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build an Agritech Crop Yield & Livestock Diagnostics Agent with LlamaIndex