Photo by Author | Canva
“AI agents will be an integral part of our daily life, which will help us in everything from scheduling appointments to our financial affairs. They will make our lives easier and more effective.”
And Andrew NG
After the growing popularity of large language models (LLMS), the next big thing is AI agent. As Andrew NG has said, they will become a part of our daily life, but how will this affect the analytical workflow? Can this be the end of manual data analytics, or can the existing workflow increases?
In this article, we tried to find out the answer to this question and analyze the timeline to find out if it was too soon or too late.
The past of data analytics
Data analytics was not as easy or fast as it is today. In fact, it went through several different stages. It is formed by the technology of its time and the increasing demand for data -powered decision -making from companies and individuals.
Microsoft Domination of Excel
In the 90s and the early 2000s, we used Microsoft Excel for everything. Remember the school assignments or tasks in your workplace. You had to collect columns and write long formulas and configure them. There are not many sources where you can learn them, so the courses are very famous.
Large datases will reduce this process, and make a report was manual and repeated.
The rise of SQL, Azigar, R
Finally, Excel began to be short. Here, SQL stepped inside. And since then it has been a rock star. This structure is expanded and sharp. You probably remember the first time you used the SQL. In seconds, he analyzed.
R was there, but with the development of Azar, it has also been increased. It is equivalent to talking with data because of its syntax. Now complex tasks can be done in minutes. The companies also saw it, and everyone was looking for capabilities that could work with SQL, Azigar and R. It was a new standard.
Two dashboards everywhere
After 2018, a new shift took place. Tools like tables and power BI just click data analysis, and they offer a wonderful concept together, called dashboards. These nine code tools have become so rapidly popular, and now all companies are changing their job details.
Powerbe or tablo experiences are essential!
Future: Entry of LLMS
Then, large models of language enter the scene, and what was the entrance door! Everyone is talking about LLMS and trying to integrate them into their workflow. You can often see article titles, “Will LLM Data replace analysts?“.
However, the first version of LLMS cannot offer automatic data analysis unless the Chat GPT code comes with the spokesperson. It was the game changer that frightened the data analysts, as it began to show that data analtics workflows could potentially be automated with just one click. How? Let’s see.
Data Exploration with LLMS
Consider this data project: Black Friday purchases. It has been used as a tech home assignment in the process of recruiting data science position in Walmart.
The link to this data project is:
See, download the data, and upload it to Chat GPT. Use this quick structure:
I have attached my dataset.
Here is my dataset description:
(Copy-paste from the platform)
Perform data exploration using visuals.Here is the first part of the output.
But it is not over yet. It’s going on, so let’s see what else we have to show.
Now we have the overall summary of datasets and concepts. Let’s look at the third part of the data research, which is now verbal.
The best part? He did all this in seconds. But AI agents are a bit more advanced than that. So, let’s create an AI agent that automatically make data search.
Data Analytical Agent
The agents went one step further by the traditional LLM interactions. As powerful they were LLM, it felt as if something was missing. Or is it an indispensable desire for humanity to discover an intelligence that exceeds their own limits? LLMS LOY, you had to point out as we did above, but they do not even need human intervention by data analtics agents. They will do everything themselves.
Data search and concept agent implementation
Let’s make an agent together. To do this, we will use Langchen and Streamlit.
Agent compile
First, let’s install all libraries.
import streamlit as st
import pandas as pd
warnings.filterwarnings('ignore')
from langchain_experimental.agents.agent_toolkits import create_pandas_dataframe_agent
from langchain_openai import ChatOpenAI
from langchain.agents.agent_types import AgentType
import io
import warnings
import matplotlib.pyplot as plt
import seaborn as snsOur streamlit agent lets you upload CSV or Excel file with this code.
api_key = "api-key-here"
st.set_page_config(page_title="Agentic Data Explorer", layout="wide")
st.title("Chat With Your Data — Agent + Visual Insights")
uploaded_file = st.file_uploader("Upload your CSV or Excel file", type=("csv", "xlsx"))
if uploaded_file:
# Read file
if uploaded_file.name.endswith(".csv"):
df = pd.read_csv(uploaded_file)
elif uploaded_file.name.endswith(".xlsx"):
df = pd.read_excel(uploaded_file)Next, data search and data visualization codes come. As you can see, there are some if Blocks that will apply your code based on the features of datases uploaded.
# --- Basic Exploration ---
st.subheader("📌 Data Preview")
st.dataframe(df.head())
st.subheader("🔎 Basic Statistics")
st.dataframe(df.describe())
st.subheader("📋 Column Info")
buffer = io.StringIO()
df.info(buf=buffer)
st.text(buffer.getvalue())
# --- Auto Visualizations ---
st.subheader("📊 Auto Visualizations (Top 2 Columns)")
numeric_cols = df.select_dtypes(include=("int64", "float64")).columns.tolist()
categorical_cols = df.select_dtypes(include=("object", "category")).columns.tolist()
if numeric_cols:
col = numeric_cols(0)
st.markdown(f"### Histogram for `{col}`")
fig, ax = plt.subplots()
sns.histplot(df(col).dropna(), kde=True, ax=ax)
st.pyplot(fig)
if categorical_cols:
# Limiting to the top 15 categories by count
top_cats = df(col).value_counts().head(15)
st.markdown(f"### Top 15 Categories in `{col}`")
fig, ax = plt.subplots()
top_cats.plot(kind='bar', ax=ax)
plt.xticks(rotation=45, ha="right")
st.pyplot(fig)Next, set an agent.
st.divider()
st.subheader("🧠Ask Anything to Your Data (Agent)")
prompt = st.text_input("Try: 'Which category has the highest average sales?'")
if prompt:
agent = create_pandas_dataframe_agent(
ChatOpenAI(
temperature=0,
model="gpt-3.5-turbo", # Or "gpt-4" if you have access
api_key=api_key
),
df,
verbose=True,
agent_type=AgentType.OPENAI_FUNCTIONS,
**{"allow_dangerous_code": True}
)
with st.spinner("Agent is thinking..."):
response = agent.invoke(prompt)
st.success("✅ Answer:")
st.markdown(f"> {response('output')}")To examine the agent
Now everything is ready. Save it like:
Next, go to the working directory of this script file, and run it using this code:
And, voila!
Your agent is ready, let’s check it out!
The final views
In this article, we have analyzed the data analtics evolution that began in the 90s, from Excel to LLM agents. We have analyzed This real -life datasetIn a real data science job interview, was asked using GPT.
Finally, we have developed an agent that automates data and automate data using streamllate, langchain, and other libraries, which is a crossroads of past and new data analtics workflow. And we did everything using a real -life data project.
Whether you adopt them today or tomorrow, AI agents are no longer in the future. In fact, they are the next step of analytics.
Net Razii A data is in a scientist and product strategy. He is also an affiliated professor of Teaching Analytics, and is the founder of Stratskrich, a platform that helps data scientists prepare for his interview with the real questions of high companies. The net carrier writes on the latest trends in the market, gives interview advice, sharing data science projects, and everything covers SQL.