Skip to main content
Loading navigation
Back to challenges

CrewAI & Bland AI Agritech DPI Advisory Agent

Smallholder farmers require rapid AI advisory services integrated with government agritech public infrastructure data. You will build a multi-agent workflow that ingests soil health cards, market pricing, and telemetry to deliver voice-based agricultural guidance via automated phone workflows.

Business OperationsHosted by Vera
Sign in free to start

What you are building

Orchestrate specialized CrewAI agents paired with Bland AI phone interfaces to process agritech public infrastructure data and automate voice advisory dispatch.

How work is evaluated

Evaluation

Validates multi-agent advisory generation and automated voice script compilation for agricultural support.

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

Call Dispatch Success

Checks that the Bland AI call dispatch returns a valid call ID

Binary check

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

Dimension 2

Agronomic Accuracy

Evaluates correctness of fertilizer ratios • target: 0.98 • range: 0.9-1

Binary check

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

Learning goals

  • Design multi-role CrewAI orchestration for soil data analysis and market linkage

  • Integrate Bland AI REST APIs for automated telephone voice calls to farmers

  • Process government Soil Health Card datasets with localized crop recommendations

  • Ensure voice conversational outputs adhere to rural dialect and language constraints

Resources

Dataset notes

Agricultural soil health test readings paired with current mandi market prices across major Indian districts.

Challenge setup

Validates multi-agent advisory generation and automated voice script compilation for agricultural support.

Challenge input

JSON with farmer ID, language, N-P-K soil values, and crop preference

CrewAI

Orchestrates complex agent roles and tasks.

Bland AI

Executes natural language AI phone calls.

Evaluated output

JSON object with recommendations and Bland AI call ID

Evaluation checks
  • Checks that the Bland AI call dispatch returns a valid call ID
  • Evaluates correctness of fertilizer ratios • target: 0.98 • range: 0.9-1
Configured evaluation evidence
  • Benchmark: AgriAgent-Eval
  • Agronomic Accuracy target: 0.98
  • 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
  • Bland AI · Required
  • CrewAI · Optional
  • Guidance · Optional

Evaluation contract

  • Call Dispatch Success · Weight 1
  • Agronomic Accuracy · Weight 1

Metrics

  • Agronomic Accuracy · Target 0.98

Benchmarks

  • AgriAgent-Eval
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON with farmer ID, language, N-P-K soil values, and crop preference

    Output: JSON object with recommendations and Bland AI call ID

    Execution harness: Python

Evaluation tools

  • CrewAI · Agent Framework
  • Bland AI · Voice Dispatch

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about CrewAI & Bland AI Agritech DPI Advisory Agent