AI Development
Advanced
Always open

Agent for Enterprise M&A Due Diligence

This challenge focuses on building an advanced RAG-powered agent using LlamaIndex for enterprise M&A due diligence, inspired by the news of cloud providers acquiring AI search companies to enhance agent capabilities. Participants will create an intelligent agent capable of querying internal and external knowledge bases to gather, synthesize, and analyze critical information pertinent to a potential acquisition target. The agent will need to handle diverse data types (documents, web pages, internal reports) and provide concise, actionable insights. The system should demonstrate sophisticated retrieval augmentation, dynamic tool selection, and the ability to answer complex, multi-hop questions about a target company's financials, market position, and technological landscape. This requires leveraging LlamaIndex's advanced indexing, query engine capabilities, and agent orchestration to ensure accurate and up-to-date information retrieval.

Challenge brief

What you are building

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

This challenge focuses on building an advanced RAG-powered agent using LlamaIndex for enterprise M&A due diligence, inspired by the news of cloud providers acquiring AI search companies to enhance agent capabilities. Participants will create an intelligent agent capable of querying internal and external knowledge bases to gather, synthesize, and analyze critical information pertinent to a potential acquisition target. The agent will need to handle diverse data types (documents, web pages, internal reports) and provide concise, actionable insights. The system should demonstrate sophisticated retrieval augmentation, dynamic tool selection, and the ability to answer complex, multi-hop questions about a target company's financials, market position, and technological landscape. This requires leveraging LlamaIndex's advanced indexing, query engine capabilities, and agent orchestration to ensure accurate and up-to-date information retrieval.

Datasets

Shared data for this challenge

Review public datasets and any private uploads tied to your build.

Loading datasets...
Evaluation rubric

How submissions are scored

These dimensions define what the evaluator checks, how much each dimension matters, and which criteria separate a passable run from a strong one.

Max Score: 4
Dimensions
4 scoring checks
Binary
4 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1jsonformatcheck

JsonFormatCheck

Verify the output is a valid JSON matching the specified schema.

binary
Weight: 1
Binary check

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

Dimension 2sourcescited

SourcesCited

Ensure at least two distinct sources are cited in the output.

binary
Weight: 1
Binary check

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

Dimension 3answeraccuracy

AnswerAccuracy

Factual correctness and completeness of the answer compared to a ground truth (human-evaluated or via automated checks for simple facts). • target: 0.9 • range: 0-1

binary
Weight: 1
Binary check

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

Dimension 4sourcerelevance

SourceRelevance

Average relevance of cited sources to the generated answer. • target: 4.5 • range: 1-5

binary
Weight: 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

Master LlamaIndex for building sophisticated RAG pipelines, including data ingestion, indexing strategies (vector, keyword, hierarchical), and query engines

Implement a LlamaIndex agent capable of dynamically selecting and using various tools (e.g., web scraping, document parsers, database connectors) for M&A-specific data retrieval

Design and configure query engines within LlamaIndex to leverage Gemini 2.5 Pro for advanced reasoning, summarization, and synthesis of retrieved information

Utilize Skyvern as a LlamaIndex tool to automate the extraction of financial data, news articles, and competitive analysis reports from public websites for due diligence

Integrate Ollama for local embedding generation and/or local LLM inference during development and testing, allowing for faster iteration and reduced API costs

Develop strategies for handling semi-structured and unstructured data, using LlamaIndex's capabilities to extract key entities and relationships relevant to M&A analysis

Build a multi-modal RAG capability within LlamaIndex to potentially process visual information (e.g., charts, infographics) alongside text-based documents for comprehensive analysis

Implement evaluation metrics and logging within LlamaIndex to monitor retrieval accuracy and agent performance during complex due diligence queries

Start from your terminal
$npx -y @versalist/cli start agent-for-enterprise-m-a-due-diligence

[ok] Wrote CHALLENGE.md

[ok] Wrote .versalist.json

[ok] Wrote eval/examples.json

Requires VERSALIST_API_KEY. Works with any MCP-aware editor.

Docs
Manage API keys
Challenge at a glance
Host and timing
Vera

AI Research & Mentorship

Starts Available now
Evergreen challenge
Your progress

Participation status

You haven't started this challenge yet

Timeline and host

Operating window

Key dates and the organization behind this challenge.

Start date
Available now
Run mode
Evergreen challenge
Explore

Find another challenge

Jump to a random challenge when you want a fresh benchmark or a different problem space.

Useful when you want to pressure-test your workflow on a new dataset, new constraints, or a new evaluation rubric.

Tool Space Recipe

Draft
Evaluation
Rubric: 4 dimensions
·JsonFormatCheck(1%)
·SourcesCited(1%)
·AnswerAccuracy(1%)
·SourceRelevance(1%)
Gold items: 1 (1 public)

Frequently Asked Questions about Agent for Enterprise M&A Due Diligence