What Makes Streamlit Different

One AI engineering post, weekly

LLM benchmarks, prompt techniques, and token-cost breakdowns — not another AI news roundup.

Streamlit converts a Python script into a web app by running it top-to-bottom every time a user interacts with a widget. No HTML, no JavaScript, no React. Just Python.

This execution model is the key to understanding Streamlit. Every button click, slider move, or text input triggers a full script rerun. This sounds inefficient, but Streamlit optimizes it with caching.

Your First Streamlit App

python

import streamlit as st
import pandas as pd

st.title("Sales Dashboard")

uploaded = st.file_uploader("Upload CSV", type="csv")
if uploaded:
    df = pd.read_csv(uploaded)

    region = st.selectbox("Filter by region", df["region"].unique())
    filtered = df[df["region"] == region]

    st.dataframe(filtered)
    st.bar_chart(filtered.groupby("product")["revenue"].sum())

Run with: streamlit run app.py

Team workspace

Ship faster with chat, meetings, and projects in one place — Zlyqor.

Start free

Caching: The Essential Optimization

python

import streamlit as st
import pandas as pd
from sentence_transformers import SentenceTransformer

@st.cache_data  # cache data loading  -  reruns only when inputs change
def load_data(path: str) -> pd.DataFrame:
    return pd.read_parquet(path)

@st.cache_resource  # cache heavy objects  -  loaded once, shared across sessions
def load_model() -> SentenceTransformer:
    return SentenceTransformer("all-MiniLM-L6-v2")

df = load_data("embeddings.parquet")  # cached after first run
model = load_model()  # loaded once, reused

Use @st.cache_data for data (serializable, per-user). Use @st.cache_resource for models and database connections (shared across all users).

Building an LLM Chatbot with st.chat_message

python

import streamlit as st
from openai import OpenAI

client = OpenAI(api_key=st.secrets["OPENAI_API_KEY"])

if "messages" not in st.session_state:
    st.session_state.messages = []

for msg in st.session_state.messages:
    with st.chat_message(msg["role"]):
        st.write(msg["content"])

if prompt := st.chat_input("Ask anything..."):
    st.session_state.messages.append({"role": "user", "content": prompt})
    with st.chat_message("user"):
        st.write(prompt)

    with st.chat_message("assistant"):
        response = client.chat.completions.create(
            model="gpt-4o-mini",
            messages=st.session_state.messages,
            stream=True,
        )
        content = st.write_stream(response)

    st.session_state.messages.append({"role": "assistant", "content": content})

Session State for Persistence

python

import streamlit as st

if "counter" not in st.session_state:
    st.session_state.counter = 0

if st.button("Increment"):
    st.session_state.counter += 1

st.write(f"Count: {st.session_state.counter}")

Session state persists across reruns for a single user session.

Secrets Management

toml

# .streamlit/secrets.toml (never commit this)
OPENAI_API_KEY = "sk-..."
DATABASE_URL = "mongodb+srv://..."

python

import streamlit as st
key = st.secrets["OPENAI_API_KEY"]

On Streamlit Community Cloud, secrets are set in the app dashboard.

Deploying for Free

Push your app to GitHub
Go to share.streamlit.io
Connect your repo and set the main file path
Add secrets in the dashboard
Deploy - public URL in seconds

Resources: Streamlit, GitHub, app gallery.

#streamlit#ml-apps#python#deployment#data-visualization

Mahmudul Haque Qudrati

CEO & ML Engineer

Visionary technologist, software engineer, and machine learning specialist. Founder and CEO of Pristren, directing engineering teams that ship production-grade AI/ML pipelines, mission-critical full-stack applications, and developer tooling. Creator of Zlyqor, the unified team workspace platform. Author of 540+ technical guides and benchmark research reports on large language models, agentic workflows, Model Context Protocol (MCP), and modern web stacks.