Added *__pycache__/ to .gitignore

.gitignore

@@ -1,2 +1,4 @@
 .env
-langchain_env/
+langchain_env/
+*__pycache__/
+config.py

app.py

@@ -1,52 +1,55 @@
-
 from config import environment_settings, connect_to_llm
 
-environment_settings()
-llm = connect_to_llm()
-results=llm.invoke("Tell, me who is the GOAT, iLya suskever or Andrej Karpathy?")
-print(llm)
 
-# print output
-print(results)
+if not environment_settings():
+    print("Exiting...")
+
+else:
+    llm = connect_to_llm()
+    results=llm.invoke("Tell, me who is the GOAT, iLya suskever or Andrej Karpathy?")
+    print(llm)
+
+    # print output
+    print(results)
 
-# A PROMPT MUST ALWAYS HAVE THIS FORMAT: ICE FC=> 
-# 1. intent
-# 2. Context
-# 3. Examples
-# 4. Format
-# 5. Constraints
-# 6. Polish
-# 7. Iaterate
-prompt="""
-INTENT:  
-You are to behave as a world-class football analyst, widely regarded as the best in the world. Always provide factual, concise, and professional responses to the user’s queries.  
+    # A PROMPT MUST ALWAYS HAVE THIS FORMAT: ICE FC=> 
+    # 1. intent
+    # 2. Context
+    # 3. Examples
+    # 4. Format
+    # 5. Constraints
+    # 6. Polish
+    # 7. Iaterate
+    prompt="""
+    INTENT:  
+    You are to behave as a world-class football analyst, widely regarded as the best in the world. Always provide factual, concise, and professional responses to the user’s queries.  
 
-CONTEXT:  
-You have 30 years of experience playing professional football and 12 years coaching at elite levels. You now work as a football analyst with a PhD in Applied Mathematics. You are meticulous, detail-oriented, and always rely on deep research from online sources, books, databases, APIs, and documents to support your answers.  
+    CONTEXT:  
+    You have 30 years of experience playing professional football and 12 years coaching at elite levels. You now work as a football analyst with a PhD in Applied Mathematics. You are meticulous, detail-oriented, and always rely on deep research from online sources, books, databases, APIs, and documents to support your answers.  
 
-FORMAT:  
-- Always start by introducing yourself in this style:  
+    FORMAT:  
+    - Always start by introducing yourself in this style:  
 
-"Mr. Gita, I was born with football, Sir/Madam.  
-I have played for clubs such as Arsenal, Chelsea, Barcelona, Real Madrid, PSG, and Manchester United.  
-I have coached teams like Liverpool, Manchester City, and Tottenham.  
-I have won the Ballon d’Or 5 times and the FIFA World Player of the Year 4 times.  
-I have also won the UEFA Champions League 3 times and the English Premier League 4 times.  
-I am considered one of the greatest football players of all time. How can I help you today?"  
+    "Mr. Gita, I was born with football, Sir/Madam.  
+    I have played for clubs such as Arsenal, Chelsea, Barcelona, Real Madrid, PSG, and Manchester United.  
+    I have coached teams like Liverpool, Manchester City, and Tottenham.  
+    I have won the Ballon d’Or 5 times and the FIFA World Player of the Year 4 times.  
+    I have also won the UEFA Champions League 3 times and the English Premier League 4 times.  
+    I am considered one of the greatest football players of all time. How can I help you today?"  
 
-- After this introduction, answer the user’s query in **less than 100 words**. Keep it short, precise, and professional.  
+    - After this introduction, answer the user’s query in **less than 100 words**. Keep it short, precise, and professional.  
 
-CONSTRAINTS:  
-- Always maintain a professional tone.  
-- Always keep answers under 100 words.  
-- Always provide factual responses.  
+    CONSTRAINTS:  
+    - Always maintain a professional tone.  
+    - Always keep answers under 100 words.  
+    - Always provide factual responses.  
 
-EXAMPLES:  
-Q: Who is the GOAT of football?  
-A: "The GOAT of football is Lionel Messi."  
+    EXAMPLES:  
+    Q: Who is the GOAT of football?  
+    A: "The GOAT of football is Lionel Messi."  
 
 
-"""
+    """
 
-for part in llm.stream(prompt):
-    print(part, end="", flush=True)
+    for part in llm.stream(prompt):
+        print(part, end="", flush=True)

config/settings.py

@@ -1,30 +1,78 @@
+import os
+from dotenv import load_dotenv, find_dotenv
+from langchain_google_genai import ChatGoogleGenerativeAI, GoogleGenerativeAI
+from langchain_groq import ChatGroq
+
 def environment_settings():
-    import os
-    from dotenv import load_dotenv, find_dotenv
-    load_dotenv(find_dotenv())
-    # print the keys
-    GOOGLE_API_KEY = os.getenv("GOOGLE_API_KEY")
-    GROQ_API_KEY = os.getenv("GROQ_API_KEY")
-    print(f"Google API Key: {GOOGLE_API_KEY}")  
-    print(f"Groq API Key: {GROQ_API_KEY}")
+
+    try:
+        # Load environment variables
+        if not load_dotenv(find_dotenv()):
+            raise EnvironmentError("No .env file found")
+
+        # Get API keys
+        GOOGLE_API_KEY = os.getenv("GOOGLE_API_KEY")
+        GROQ_API_KEY = os.getenv("GROQ_API_KEY")
+
+        # Validate API keys
+        missing_keys = []
+        if not GOOGLE_API_KEY:
+            missing_keys.append("GOOGLE_API_KEY")
+        if not GROQ_API_KEY:
+            missing_keys.append("GROQ_API_KEY")
+
+        if missing_keys:
+            raise ValueError(f"Missing required environment variables: {', '.join(missing_keys)}")
+
+        return True
+
+    except ValueError as ve:
+        print(f"Environment Error: {ve}")
+        print("Please check your .env file and ensure all required API keys are set.")
+        return False
+    except EnvironmentError as ee:
+        print(f"Environment Error: {ee}")
+        print("Please make sure .env file exists in the project root directory.")
+        return False
+    except Exception as e:
+        print(f"Unexpected error: {e}")
+        return False
 
 def connect_to_llm():
-    from langchain_google_genai import GoogleGenerativeAI
+    environment_settings()
     llm=GoogleGenerativeAI(model="gemini-2.5-flash", temperature=0.7)
     return llm
 def connect_to_llm_chat():
-    from langchain_google_genai import ChatGoogleGenerativeAI
+    environment_settings()
     llm=ChatGoogleGenerativeAI(model="gemini-2.5-flash", temperature=0.7)
     return llm
 def connect_to_groq():
-    from langchain_groq import ChatGroq
+    environment_settings()
     llm=ChatGroq(model="deepseek-r1-distill-llama-70b", temperature=0.7)
     return llm
 
 def connect_to_db_url():
-    import os
-    from dotenv import load_dotenv, find_dotenv
     load_dotenv(find_dotenv())
     DATABASE_URL = os.getenv("DATABASE_URL")
     print(f"Database URL: {DATABASE_URL}")
-    return DATABASE_URL
+    return DATABASE_URL
+
+def select_model():
+    print("\nAvailable Models:")
+    print("1. Google LLM")
+    print("2. Google Chat Model")
+    print("3. Groq Chat Model")
+
+    while True:
+        try:
+            choice = int(input("\nSelect a model (1-3): "))
+            if choice == 1:
+                return connect_to_llm()
+            elif choice == 2:
+                return connect_to_llm_chat()
+            elif choice == 3:
+                return connect_to_groq()
+            else:
+                print("Invalid choice. Please select 1, 2, or 3.")
+        except ValueError:
+            print("Please enter a valid number.")

configs.py

@@ -0,0 +1,36 @@
+# config.py
+import os
+from config import connect_to_llm, connect_to_llm_chat, connect_to_groq
+
+def load_google_llm():
+    llm = connect_to_llm()
+    return llm
+
+def load_google_chat_model():
+    chat_model = connect_to_llm_chat()
+    return chat_model
+
+def load_groq_chat_model():
+    groq_model = connect_to_groq()
+    return groq_model
+
+def select_model():
+    print("\nAvailable Models:")
+    print("1. Google LLM")
+    print("2. Google Chat Model")
+    print("3. Groq Chat Model")
+
+    while True:
+        try:
+            choice = int(input("\nSelect a model (1-3): "))
+            if choice == 1:
+                return load_google_llm()
+            elif choice == 2:
+                return load_google_chat_model()
+            elif choice == 3:
+                return load_groq_chat_model()
+            else:
+                print("Invalid choice. Please select 1, 2, or 3.")
+        except ValueError:
+            print("Please enter a valid number.")
+

exercises/exercise_1/README.md

@@ -0,0 +1,6 @@
+# Exercise 1
+
+## Notes
+
+- The `config` folder and its contents were used to complete this task, not the `config.py` file.
+- The model selection function was implemented in the `settings.py` file.

exercises/exercise_2/README.md

@@ -0,0 +1,31 @@
+# Exercise 2: Your First LLM Completion
+
+**Objective**: Use a completion model to generate text responses
+
+**Concepts**: LLM completion, Basic prompting, Streaming responses
+
+### Instructions
+
+1. Load a Google completion model using your config
+2. Create a simple prompt about a topic you're interested in
+3. Get a response using the `invoke()` method
+4. Implement streaming to see the response generated in real-time
+
+### Starter Code
+
+```python
+from config import load_google_llm
+
+# TODO: Load the model
+# TODO: Create a prompt
+# TODO: Get response using invoke()
+# TODO: Implement streaming with stream()
+```
+
+### Expected Behavior
+- See immediate response with invoke()
+- Watch text appear word-by-word with streaming
+- Handle the response content properly
+
+### Challenge
+Create a simple loop that accepts user prompts and responds using the completion model.

exercises/exercise_2/solution.py

@@ -0,0 +1,15 @@
+from configs import load_google_llm
+import time
+
+llm = load_google_llm()
+while True:
+    prompt = input("\nEnter your prompt: ")
+
+    # response = llm.invoke(prompt)
+    # print("Response from invoke:\n", response)
+
+    # time.sleep(4)
+
+    print("\nResponse from stream:")
+    for part in llm.stream(prompt):
+        print(part, end="", flush=True)

exercises/exercise_3/README.md

@@ -0,0 +1,34 @@
+# Exercise 3: Basic Chat Model Interaction
+
+**Objective**: Build conversational interactions using chat models
+
+**Concepts**: Chat models, Message formatting, System prompts, Conversation flow
+
+### Instructions
+
+1. Load a Google chat model from your config
+2. Create a message structure with system and user roles
+3. Use the chat model to respond to questions
+4. Experiment with different system prompts to change AI behavior
+
+### Starter Code
+
+```python
+from config import load_google_chat_model
+
+# TODO: Load chat model
+# TODO: Create messages array with system and user messages
+# TODO: Get response using invoke()
+# TODO: Test different system prompts
+```
+
+### Expected Message Format
+```python
+messages = [
+    ("system", "You are a helpful assistant specialized in..."),
+    ("user", "Your question here")
+]
+```
+
+### Challenge
+Create different personas by changing the system message and test how responses change.

exercises/exercise_3/solution.py

@@ -0,0 +1,9 @@
+from configs import load_google_chat_model
+
+llm_chat = load_google_chat_model()
+messages = [
+    ("system", "You are a helpful assistant specialized in human anatomy"),
+    ("user", "Can digestive acid in my stomach be secreted even without food in it?")
+]
+response = llm_chat.invoke(messages)
+print(f"Assistant: {response}")

exercises/exercise_4/README.md

@@ -0,0 +1,29 @@
+# Exercise 4: Streaming Responses
+
+## Goal
+
+The goal of this exercise is to demonstrate how to stream responses from a language model. This is useful when you want to display the response to the user as it is being generated, rather than waiting for the entire response to be available.
+
+## Steps
+
+1.  **Import necessary libraries**: You will need to import the `connect_to_llm_chat` function from the `config` module.
+2.  **Connect to the language model**: Use the `connect_to_llm_chat` function to get a chat model object.
+3.  **Create a loop**: Create a `while True` loop to continuously get user input.
+4.  **Use the `stream` method**: Inside the loop, call the `stream` method on the chat model object, passing the user's prompt as an argument.
+5.  **Print the response**: The `stream` method returns a generator. Iterate over the generator and print the `content` of each part of the response. Use `end=""` and `flush=True` in the `print` function to ensure the response is printed as it is received.
+
+## Solution
+
+```python
+from config import connect_to_llm_chat
+
+chat_model = connect_to_llm_chat()
+
+while True:
+    prompt = input("User prompt: ")
+    # Streaming pattern
+    print("\nBot:")
+    for part in chat_model.stream(prompt):
+        print(part.content, end="", flush=True)
+```
+

exercises/exercise_4/solution.py

@@ -0,0 +1,10 @@
+from configs import load_google_chat_model
+
+chat_model = load_google_chat_model()
+
+while True:
+    prompt = input("User prompt: ")
+    # Streaming pattern
+    print("\nBot:")
+    for part in chat_model.stream(prompt):
+        print(part.content, end="", flush=True)

exercises/exercise_5/README.md

@@ -0,0 +1 @@
+

exercises/exercise_5/solution.py

@@ -0,0 +1,14 @@
+from configs  import load_google_chat_model
+
+chat = load_google_chat_model
+
+print("THE AI ASSISTANT WELCOMES YOU")
+print("-"*30,"\n Begin by entering a prompt")
+
+while True:
+    prompt = input("USER: ")
+    if prompt.lower() == "quit" or "exit":
+        print("Goodbyte...")
+        break
+    for part in chat.stream(prompt):
+        print(part, end="", flush=True)

exercises/exercise_6/README.md

@@ -0,0 +1 @@
+