Created my first version of the AI chatbot

getting-started/test_setup.py

@@ -0,0 +1,22 @@
+import sys
+print(f"Python version: {sys.version}")
+
+try:
+    from google.cloud import aiplatform
+    print("✅ google-cloud-aiplatform installed")
+except ImportError:
+    print("❌ google-cloud-aiplatform not found")
+
+try:
+    import streamlit
+    print("✅ streamlit installed")
+except ImportError:
+    print("❌ streamlit not found")
+
+try:
+    import dotenv
+    print("✅ python-dotenv installed")
+except ImportError:
+    print("❌ python-dotenv not found")
+
+print("\n✅ All packages installed successfully!")

requirements.txt

@@ -8,6 +8,9 @@
 google-generativeai>=0.3.0
 python-dotenv>=1.0.0
 streamlit>=1.28.0
+google-cloud-aiplatform>=1.26.0
+vertexai>=0.1.0
+requests>=2.31.0
 
 # ============================================================================
 # Optional: Alternative Platforms (Uncomment to use)

sessions/session-01-ai-chatbots/live-demo/wcc_app.py

@@ -30,7 +30,7 @@
     pass
 
 # Configure Gemini API (done once at startup)
-api_key = os.getenv('GEMINI_API_KEY') or os.getenv("GOOGLE_API_KEY")
+api_key = os.getenv('API_KEY') or os.getenv("API_KEY")
 if api_key:
     genai.configure(api_key=api_key)
     print(f"✓ API configured: {bool(api_key)}")
@@ -372,16 +372,16 @@ def step_5_streamlit_interface():
     # Each step builds on the previous one, showing clear code differences!
 
     # Step 1: Basic API integration with Gemini
-    # step_1_basic_api_call()
+    step_1_basic_api_call()
 
     # Step 2: AI personality with system prompts
-    # step_2_add_personality()
+    step_2_add_personality()
 
     # Step 3: Conversation memory management
-    # step_3_conversation_memory()
+    step_3_conversation_memory()
 
     # Step 4: Model parameter experimentation
-    # step_4_model_parameters()
+    step_4_model_parameters()
 
     # Step 5: Web interface with Streamlit (with personality + conversation history)
     step_5_streamlit_interface()  # Run with: streamlit run wcc_app.py

sessions/session-01-ai-chatbots/live-demo/wcc_demo.py

@@ -19,7 +19,7 @@
 import google.generativeai as genai
 from datetime import datetime
 
-
+#this sets the model to gemini 2.5
 MODEL_ID = 'gemini-2.5-flash-lite'
 
 # Load .env if available (dev convenience)
@@ -30,9 +30,10 @@
     pass
 
 # Configure Gemini API (done once at startup)
-api_key = os.getenv('GEMINI_API_KEY') or 'your-gemini-api-key-here'
+api_key = os.getenv('API_KEY') or "API_KEY"
 genai.configure(api_key=api_key)
 
+
 print("🚀 WCC AI Learning Series - Session 1 Demo")
 print("=" * 50)
 
@@ -146,25 +147,34 @@ def step_3_conversation_memory():
     print("STEP 3: Adding Conversation Memory")
     print("=" * 60)
 
+    #defines the new class
     class WCCChatBot:
         """Chatbot with conversation memory"""
-
+        #the init is the constructor, which is the first thing that runs, whenver a new object is created
         def __init__(self):
+            #self.variable name is an instance variable, unique data that only belongs to one object
+            #creates the brain of the chatbot and saves it inside self.model
             self.model = genai.GenerativeModel(
+                #includes the model's ID
                 MODEL_ID,
+                #the system prompt needed to instruct the model on what to do
                 system_instruction=WCC_SYSTEM_PROMPT
             )
+            #this instance variable creates an empty list, this will act as the memory 
             self.conversation_history = []
-
+        #the chat method, is an instance method, a function that the object can perform
+        # is must take self as the first argument, then user input as the second variable
         def chat(self, user_input):
             """Send message and get response with context"""
+            # this updates the memory with the user input
             # Add user message to history
             self.conversation_history.append({"role": "user", "parts": [user_input]})
-
+            # creates an instance variable that passes the history to the model to generate a response
             # Generate response with full history
             response = self.model.generate_content(self.conversation_history)
 
-            # Add assistant response to history
+            # the response is then added to the model's history
+            # Add assistant response to history, role indicates who said the message, parts indicates the context
             self.conversation_history.append({"role": "model", "parts": [response.text]})
 
             return response.text
@@ -230,6 +240,7 @@ def step_4_model_parameters():
         generation_config = genai.types.GenerationConfig(
             temperature=temp,
             max_output_tokens=100,
+            top_p=0.9
         )
 
         model_temp = genai.GenerativeModel(
@@ -364,13 +375,14 @@ def step_5_streamlit_interface():
     step_1_basic_api_call()
 
     # Step 2: AI personality with system prompts
-    # step_2_add_personality()
+    step_2_add_personality()
+
 
     # Step 3: Conversation memory management
-    # step_3_conversation_memory()
+    step_3_conversation_memory()
 
     # Step 4: Model parameter experimentation
-    # step_4_model_parameters()
+    step_4_model_parameters()
 
     # Step 5: Web interface with Streamlit
-    # step_5_streamlit_interface()  # Run with: streamlit run wcc_demo.py
+    step_5_streamlit_interface()  # Run with: streamlit run wcc_demo.py

sessions/session-01-ai-chatbots/participants/sharmin-osoo/code/README.md

@@ -0,0 +1,84 @@
+Welcome! This **AI Chatbot** is your specialized mini-assistant for networking, community building, and finding relevant professional events near you.
+
+## ✨ Project Overview
+
+The **Event & Community Finder** is a conversational tool designed to streamline the process of finding local, relevant events from platforms like Meetup and Eventbrite.
+
+### 🎯 The Core Functionality
+
+The chatbot is specifically engineered to gather three key pieces of information from the user:
+
+1.  **📍 Location:** Where the user is.
+2.  **🗓️ Availability:** Which days the user is free.
+3.  **📚 Focus Area:** Whether they seek **Tech**, **Business**, or **Career-focused** events (with the ability to refine further, e.g., 'Tech events focused on AI agents').
+
+Once the inputs are gathered, the chatbot searches external platforms for available events.
+
+### 💡 Expected Output
+
+The assistant returns a short, highly useful list of matches including:
+Market size and potential Market size in the next 3-5 years
+– Competitors (both big and small)
+– Common customer pain points
+– Current market conditions & Feasibility
+– Recommendations on utilising Technology to stay ahead and Systems for efficiency
+– Raising Capital / Funding (local perspective, most suitable way based on their scenario) - this is optional and only return if the user requests
+
+# How to set up 
+
+### 1. Install Dependencies
+
+```bash
+pip install -r code/requirements.txt
+```
+
+### 2. Set Up Your API Key
+
+Create a `.env` file in the project root:
+
+```bash
+GEMINI_API_KEY=your-gemini-api-key-here
+```
+
+Get your API key: [Gemini API Key Setup Guide](../../getting-started/gemini-api-key-setup.md)
+
+
+#### what is included:
+chatbot.py - building of the chatbot, where we call the api, create conversation memory, and add parameters
+requirements.txt - ptyhon dependencies
+.env - where the API key is kept
+
+
+### 3. Run the Demo
+
+```bash
+python wcc_demo.py
+```
+
+
+# How to run it 
+
+
+
+# What i learned
+
+ Model Parameters: Explored the impact of parameters like temperature (controlling randomness and creativity), and the optimal use of Top P for token selection (e.g., setting probabilities to equal 85%).
+ Object-Oriented Programming (OOP): Had the opportunity to refresh and apply OOP principles in the code structure.
+# Challenges faced
+Initial System Prompt: The first system prompt was aiming to output a table which did not work so well when it was formatted, which made the response confusing. 
+Re-acquaintance: Getting back into the rhythm of Python coding after a break, and integrating with Streamlit for the first time as the chat interface.
+
+# Future improvements 2.0
+
+Enhanced UI/UX: Create a more intuitive and user-friendly interface.
+
+Context Window Integration: Implement a context window for the AI model to access external, rich information, providing more tailored and well-rounded responses.
+
+Performance: Optimize the model's setup to potentially reduce thinking/response time.
+
+Feature Expansion: Add functionality for users to save preferences to a file or subscribe to weekly digests.
+
+Framework Suggestions: Integrate the ability for the AI to provide relevant professional frameworks alongside event suggestions.
+
+
+