Skip to main content
Back to challenges
Challenge

Build a DeepSeek-V3 Multi-Agent Market Insight System

Inspired by the need for robust forecasting, this challenge involves developing a multi-agent system capable of generating comprehensive market insights. Participants will design and implement a collaborative team of AI agents, each specializing in a different aspect of market analysis (e.g., data gathering, trend analysis, strategic recommendation). The system must leverage advanced Retrieval-Augmented Generation (RAG) techniques, using DeepSeek-V3 for nuanced understanding. Efficient tool selection will be crucial for agents to interact with external data sources and generate actionable insights. The core task is to create an agentic AI system that can autonomously analyze a given market query, retrieve relevant data, identify trends, and synthesize a coherent report. The agents should engage in multi-role debating and multi-path reasoning to refine their findings and overcome potential biases, mirroring the sophisticated collaboration observed in human expert teams. This challenge directly addresses the growing demand for enterprise-grade agentic AI systems that can deliver high-quality, actionable intelligence.

Special Purpose AgentsHosted 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 a multi-agent system using DeepSeek-V3, LangChain, and Redis Vector for RAG to collaboratively provide comprehensive market insights with efficient tool integration.

Delivery guide

How work is evaluated

Evaluation

The evaluation will assess the multi-agent system's ability to accurately, comprehensively, and collaboratively generate market insights from a given query.

Datasets

Challenge datasets

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

Learning goals

What you should walk away with

  • Implement advanced RAG pipelines using DeepSeek-V3 and Redis Vector.

  • Develop strategies for multi-path reasoning and agent-to-agent debate.

  • Design a robust multi-agent architecture for collaborative insight discovery.

  • Orchestrate multi-agent workflows with tools for real-world data integration.

  • Evaluate the quality and accuracy of AI-generated market insights.

Resources and assets

Reference links and supporting material

Dataset notes

Sample data for 1 tasks

Challenge setup

The evaluation will assess the multi-agent system's ability to accurately, comprehensively, and collaboratively generate market insights from a given query.

Challenge input

{ "query": "string", "market_data_urls": ["string"] }

Agent execution

No agent tool is listed for this challenge.

Evaluated output

{ "report_title": "string", "summary": "string", "key_trends": [{"name": "string", "description": "string", "impact": "string"}], "opportunities":...

Evaluation checks
  • Ensures the output is a well-formed JSON object with all specified top-level keys.
  • Verifies a sample of key facts/trends against provided or external ground truth data sources (e.g., by comp...
  • Checks if 'agent_reasoning_paths' contains distinct contributions from at least three different agent roles...
Configured evaluation evidence
  • InsightDepthScore
Configured runtime
  • Python execution harness

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

No tool records are attached.

Evaluation contract

  • Ensures the output is a well-formed JSON object with all specified top-level keys.
  • Verifies a sample of key facts/trends against provided or external ground truth data sources (e.g., by comparing extracted entities against actual reports).
  • Checks if 'agent_reasoning_paths' contains distinct contributions from at least three different agent roles, indicating collaboration.

Metrics

  • InsightDepthScore
  • CoherenceScore · Target 0.8
  • ToolUtilizationRate · Target 0.75

Workflow tasks

  1. Task 1

    Input: { "query": "string", "market_data_urls": ["string"] }

    Output: { "report_title": "string", "summary": "string", "key_trends": [{"name": "string", "description": "string", "impact": "string"}], "opportunities": ["string"], "risks": ["string"], "supporting_data_sources": ["string"], "agent_reasoning_paths": ["string"] }

    Execution harness: Python

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Build a DeepSeek-V3 Multi-Agent Market Insight System