Scribd
Upload
56 views

Generative AI Engineer Assignment

Uploaded by

aryan.10119011621
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
56 views

Generative AI Engineer Assignment

Uploaded by

aryan.10119011621
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Market Research & Use Case Generation Agent

Your goal is to design a Multi-Agent architecture system that generates relevant AI and
Generative AI (GenAI) use cases for a given Company or Industry. The system will conduct
market research, understand the industry and product, and provide resource assets for AI/ML
solutions, focusing on enhancing operations and customer experiences.

The Multi-agent architecture must have the following agents:

● Research the Industry or the Company


○ Use a Web browser tool and understand the industry and segment the company is
working in (e.g., Automotive, Manufacturing, Finance, Retail, Healthcare, etc.).
○ Identify the company’s key offerings and strategic focus areas (e.g., operations,
supply chain, customer experience, etc.). A vision and product information on the
industry should be fine as well.

● Market Standards & Use Case Generation


○ Based on the industry conducted, analyze industry trends and standards within the
company’s sector related to AI, ML, and automation.
○ Propose relevant use cases where the company can leverage GenAI, LLMs, and
ML technologies to improve their processes, enhance customer satisfaction, and
boost operational efficiency.

● Resource Asset Collection


○ Collect the above use cases generated and search for relevant datasets on
platforms like Kaggle, HuggingFace, and GitHub.
○ Save the resource links fetched in a text or markdown file.
○ [OPTIONAL: BONUS] If applicable, propose GenAI solutions like document
search, automated report generation, and AI-powered chat systems for internal or
customer-facing purposes.

Final Proposal

● List down the top use cases that can be delivered to the customer, ensuring they are
relevant to the company’s goals and operational needs.
○ Ensure to add references through which certain use cases were suggested
○ Resource Asset links should be clickable.
Additional Resources:

● You may recommend the agents to refer to reports and insights on AI and digital
transformation from industry-specific sources such as McKinsey, Deloitte, or Nexocode.
● Search for industry-specific use cases (e.g., “how is the retail industry leveraging AI and
ML” or “AI applications in automotive manufacturing”).

Deliverables
● Source code: Submit the source code of your Multi-Agentic workflow, along with any
necessary scripts, configuration files, and documentation.
● Report: Write a report that explains your methodology, results, and conclusions. Include
anything that helps illustrate your findings. Architecture flowchart of your
implementation is MUST.
● Demo Video: A video showcasing the architecture flowchart and the output generated.
Bonus: Deploy the application on Streamlit or Gradio or any other relevant tools.

Assignment Evaluation Criteria:

● Your agent will be evaluated based on the following parameters:


○ Depth of market research and competitor analysis.
○ Relevance and creativity of use cases generated for the customer.
○ Feasibility and practicality of implementing the proposed solutions.
○ Quality and clarity of the final proposal, including actionable insights.

Suggested Tools & Frameworks

● Agent Frameworks: Use OpenAGI, BeyondLLM, LangChain, LLamaIndex, Autogen, or


CrewAI for building AI agents. You can check DeepLearning AI short courses for quick
tutorials or RAG & Agents Bootcamp.
● Search & Retrieval:
○ Leverage Tavily or Exa or Serper for advanced web search
● Dataset Sources: Find relevant datasets on Kaggle, HuggingFace, or GitHub etc.

Feel free to use any other tools you're comfortable with.

Learning Resources
● Deep learning Short sources:
○ https://www.deeplearning.ai/short-courses/multi-ai-agent-systems-with-crewai/
○ https://www.deeplearning.ai/short-courses/ai-agents-in-langgraph/
● RAG Agents Bootcamp: https://aiplanet.com/courses/rag-agents-bootcamp
● AutoGen Documentation detailed docs:
https://microsoft.github.io/autogen/docs/Getting-Started/
Sample Output Reports:
● Market research (only selected links are added); the competitors' list and their annual
reports could be fetched (even here, we added selected links); this allows one to tell what
industry leaders are doing wrt AI adoption
● Use Case Proposal
● Feasibility check for datasets and building use cases

You might also like