Data Analysis with LLM Agents

AI Tech Circle

Stay Ahead in AI with Weekly AI Roundup; read and listen on AITechCircle :

Welcome to the weekly AI Newsletter, where I provide actionable ideas and tips to assist you in your job and business.

Before we start, share this week's updates with a friend or a colleague:

Today at a Glance:

• Simple use case of AI Agent as Data Analyst
• Generative AI Usecase: Code generation with Large Language Models
• AI Weekly news and updates, Favorite Tip Of The Week & Things to Know
• Open Tech Talk Podcast, the latest episode on Building an AI-Driven Business with Gray Mabry

Your Data Analyst AI Agent

AI Agents are among the top ways to utilize the LLM in business. They equip you with practical tools to navigate the AI landscape. An AI agent designed for data analysis can perform tasks such as querying databases, running data processing scripts, generating visualizations, and providing insights based on the analyzed data.

This article, 'AI Agents: The Future of Enterprise AI… or Automation Repackaged?' , written by Umang Varma , already covered the basic concepts.

Based on the cookbook provided by Cohere , here's what I did to practice an AI agent for data analysis.

It connects Cohere’s Command models to external tools such as search engines, APIs, databases, and other software tools. Similar to how Retrieval-Augmented Generation (RAG) enables a model to utilize an external data source for improving factual generation, tool use allows retrieving data from multiple sources.

The AI Agent was built in this notebook to create a simple data analyst capable of searching the web and running code in a Python interpreter. This agent leverages Cohere’s Command R+ mode and Langchain to perform its tasks efficiently.

I have set up my environment by installing:

! pip install --quiet langchain langchain_cohere langchain_experimental

After this, the connection to the Cohere API is made, and the Cohere chat model is created.

Then, the agent needs to be able to search the web; to achieve this, Tavily AI is used. I created the account and got the API key.

Now, we need to have a Python interpreter. The complete notebook is available here .

Now, let's ask our Data analyst to complete the task.

Parameter passed:

"Create a plot of the top Generative AI use cases the organizations are considering for implementation."

The output:

Another Parameter:

"Create a plot of the number of full-time employees at the three tech companies with the highest market cap in the United States in 2024."

Then, it summarizes the information so that the plot can be created.

The output:

If you want fully managed AI agents, Oracle has released the OCI Generative AI Agents . This fully managed service combines the power of large language models (LLMs) with an intelligent retrieval system to create contextually relevant answers by searching your knowledge base, making your AI applications intelligent and efficient. It is in beta, and you can sign up to test your subsequent use case.

Weekly News & Updates...

Last week's AI breakthroughs marked another leap forward in the tech revolution.

1. Voice Isolator : Remove unwanted background noise and extract clear dialogue from any audio for your podcast, interview, or film sound.
2. The built-in pro mpt generator by Claude 3.5 Sonnet enables you to describe your task, and LLM will create the prompt for you.

The Cloud: the backbone of the AI revolution

• Oracle Announces Exadata Exascale, the World’s Only Intelligent Data Architecture for the Cloud
• Transforming the Developer Experience for Every Engineering Role, link
• Mission NIMpossible: Decoding the Microservices That Accelerate Generative AI, link

Gen AI Use Case of the Week:

Generative AI use cases for Tech folks:

Marketing Content AI Assistant for Content Generation with Large Language Models

Favorite Tip Of The Week:

Here's my favorite resource of the week.

• Go through this Colab notebook by Fran?ois Chollet that shows you how to build a chatbot using Gemma 2 9B

Potential of AI

• Demystifying LLM-based Software Engineering Agents , a paper outlining the new agentless approach to solve software development problems automatically.

Things to Know...

The Department of Homeland Security has published a report on reducing risks at the intersection of AI and CBRN (Chemical, Biological, Radiological, and Nuclear) threats.

The 9 findings were documented, and against each finding, there is a very detailed recommendation, which can be followed as a starting point.

The Opportunity...

Podcast:

• This week's Open Tech Talks episode 139 is "Building an AI-Driven Business with Gray Mabry." Gray Mabry, Co-Founder & CEO, Founder, iVenture Solutions has over two decades of leadership experience.

Apple | Spotify | Youtube

Courses to attend:

• Understanding Large Language Models: Foundations and Safety, a course from UC Berkeley. Course Site , Youtube

Events:

• GITEX GLOBAL , Oct 14-18, 2024, Dubai, UAE
• EUROPEAN Conference on Artificial Intelligence, Oct 19-24, 2024 Santiago de Compostela

Tech and Tools...

• Tabby is a self-hosted AI coding assistant offering an open-source and on-premises alternative to GitHub Copilot.
• AutoGPT is a generalist LLM-based AI agent that can autonomously accomplish minor tasks.

Data Sets...

• Measuring Mathematical Problem Solving With the MATH Dataset . This is the repository for Measuring Mathematical Problem Solving With the MATH Dataset.
• FEVEROUS (Fact Extraction and VERification Over Unstructured and Structured information) is a fact verification dataset which consists of 87,026 verified claims

Other Technology News

Want to stay on the cutting edge?

Here's what else is happening in Information Technology you should know about:

• OpenAI Scale Ranks Progress Toward ‘Human-Level’ Problem Solving reported by Bloomberg
• OpenAI working on new reasoning technology under code name 'Strawberry' covered by Reuters

That's it!

As always, thanks for reading.

Hit reply and let me know what you found most helpful this week - I'd love to hear from you!

Until next week,

Kashif Manzoor

The opinions expressed here are solely my conjecture based on experience, practice, and observation. They do not represent the thoughts, intentions, plans, or strategies of my current or previous employers or their clients/customers. The objective of this newsletter is to share and learn with the community.