Skip to main content
Loading navigation
Back to challenges

Autonomous EV Diagnostics with Gemini Pro

Develop an autonomous agent using Gemini Pro and Multi-Agent (A2A) to diagnose and suggest repairs for electric vehicles based on error codes and sensor data. The agent should leverage external APIs (e.g., vehicle diagnostic databases) and create comprehensive repair reports. This challenge focuses on building a tool-using agent that can autonomously gather information, reason about the problem, and propose solutions. The final output should be a detailed report suitable for a mechanic.

Special Purpose AgentsHosted by Vera
Sign in free to start

What you are building

Build an autonomous diagnostic agent for electric vehicles using Gemini Pro, Agent 2 Agent, and external APIs.

How work is evaluated

Evaluation

Evaluate the accuracy, completeness, and clarity of the generated repair reports.

Datasets

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

Learning goals

  • Build an autonomous agent using Llama 2 and LangChain.

  • Develop a system for reasoning and problem-solving based on diagnostic information.

  • Implement API calls to access external vehicle diagnostic data.

  • Generate comprehensive repair reports in a human-readable format.

Resources

Dataset notes

Sample data for 1 tasks

Challenge setup

Evaluate the accuracy, completeness, and clarity of the generated repair reports.

Challenge input

JSON containing vehicle error codes and sensor data.

Agent execution

No agent tool is listed for this challenge.

Evaluated output

Human-readable repair report (text format).

Evaluation checks
  • Checks if the report identifies the correct issues.
  • Checks if the report includes all necessary information.
Configured evaluation evidence
  • Accuracy target: 90
Configured runtime
  • Python execution harness

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

View technical recipe

Configured tools

No tool records are attached.

Evaluation contract

  • Checks if the report identifies the correct issues.
  • Checks if the report includes all necessary information.

Metrics

  • Accuracy · Target 90
  • Completeness · Target 95

Workflow tasks

  1. Task 1

    Input: JSON containing vehicle error codes and sensor data.

    Output: Human-readable repair report (text format).

    Execution harness: Python

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Autonomous EV Diagnostics with Gemini Pro