Gen-AI Developer Classroom notes 22/Jul/2026

User interface

• Options:
  • RAG will be invoked as api from chat interface (javascript)
  • POC:
    • Streamlit
    • Gradio

Streamlit

• Refer Here for the streamlit docs

• Add a package

uv add streamlit

• create a file called as app.py and to run use streamlit run app.py

• use playground

• Streamlit session state: Refer Here

Session state

using local values vs session state keys

• local value: the count will always be 1 after n clicks on button as entire code is running

import streamlit as st

count = 0
if st.button("Increment"):
    count += 1

st.write(f"Count: {count}")

• session state key: Rather than using a local variable we are using session state key

import streamlit as st

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

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

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

Widget’s value: plain variable vs its key

• plain variable

import streamlit as st

name = st.text_input("Name")
st.write(f"Hello {name}!")

• key

import streamlit as st

st.text_input("Name", key="name")
st.write(f"Hello {st.session_state.name}")

def greet():
    st.write(f"Call back sees: {st.session_state.name}")

st.button("Greet", on_click=greet)

Caching:

• To avoid recomputing expensive resources like db or files or llm streamlit supports caching Refer Here

• It has

  • cache data
  • cache resource

• Refer Here for the changes done