THE
Generative AI
Practitioner

# Variables store values — Python figures out the type automatically
name     = "Muhammad Rustam"   # string (text)
age      = 28                  # integer (whole number)
rating   = 4.8                 # float (decimal)
enrolled = True                 # boolean

# f-strings — the cleanest way to format text (essential for AI prompts!)
print(f"Hello, {name}! Your rating is {rating}/5.0")

# Lists — store multiple items
ai_tools = ["ChatGPT", "Claude", "Gemini", "Llama"]
ai_tools.append("Mistral")
print(ai_tools[0])   # ChatGPT (index starts at 0)

# For loops
for tool in ai_tools:
    print(f"Tool: {tool}")

# If / elif / else
score = 85
if score >= 90:
    grade = "A"
elif score >= 80:
    grade = "B"
else:
    grade = "Fail"
print(f"Grade: {grade}")  # Grade: B

# Define reusable functions
def ask_ai(question, model="llama3"):
    """Send a question to an AI model and return the answer."""
    try:
        response = client.chat.completions.create(
            model=model,
            messages=[{"role": "user", "content": question}]
        )
        return response.choices[0].message.content
    except Exception as e:
        print(f"Error: {e}")
        return "Could not get a response. Please try again."

answer = ask_ai("What is the capital of Pakistan?")
print(answer)  # Islamabad

Classify the sentiment of this text as Positive, Negative, or Neutral:
"The new AI course from AiBytec is absolutely amazing!"

Classify reviews as Positive, Negative, or Neutral.

Examples:
"Best laptop I've owned." → Positive
"Terrible battery life."  → Negative
"Arrived on time."         → Neutral

Now classify:
"Content is deep but platform loads slowly." →

A company has 3 AI developers. Each completes 4 projects/month.
They earn PKR 25,000 per project.
To triple monthly revenue, how many developers do they need?

Think through this step by step before giving your final answer.

system_prompt = """
You are EduAI, an intelligent tutoring assistant for AI By Tech Academy.

Your role:
- Help students understand Generative AI concepts clearly
- Explain technical topics using simple, relatable examples from Pakistan
- When students share code, debug it and explain the fix
- Encourage students and celebrate their progress

Constraints:
- Always respond in English with occasional Urdu phrases for warmth
- Keep explanations under 200 words unless the student asks for more
- Never write complete assignments — guide students to the answer
- If you don't know something, say so clearly

Tone: Friendly, encouraging, teacher-like, culturally aware
"""

# ❌ WRONG — Never hardcode API keys in your code
client = OpenAI(api_key="sk-abc123...")  # Anyone who sees your code can use your key!

# ✅ CORRECT — Use environment variables
import os
from dotenv import load_dotenv

load_dotenv()  # Loads from a .env file
client = OpenAI(api_key=os.getenv("OPENAI_API_KEY"))

# .env file (add to .gitignore, NEVER commit to GitHub):
# OPENAI_API_KEY=sk-abc123...
# GROQ_API_KEY=gsk_abc123...

from groq import Groq
import os

client = Groq(api_key=os.getenv("GROQ_API_KEY"))

def fast_ai(prompt, model="llama3-8b-8192"):
    response = client.chat.completions.create(
        model=model,
        messages=[{"role": "user", "content": prompt}],
        temperature=0.7, max_tokens=1024
    )
    return response.choices[0].message.content

answer = fast_ai("What are the top 5 AI skills for Pakistani developers in 2025?")
print(answer)

conversation_history = []

def chat(user_input, system="You are a helpful AI assistant."):
    if not conversation_history:
        conversation_history.append({"role": "system", "content": system})
    conversation_history.append({"role": "user", "content": user_input})
    response = client.chat.completions.create(
        model="llama3-8b-8192", messages=conversation_history
    )
    ai_reply = response.choices[0].message.content
    conversation_history.append({"role": "assistant", "content": ai_reply})
    return ai_reply

print(chat("My name is Rustam and I am from Karachi."))
print(chat("What is my name and where am I from?"))  # AI remembers!

from transformers import pipeline

# TEXT GENERATION
generator = pipeline("text-generation", model="gpt2")
result = generator("The future of AI in Pakistan will", max_length=100)
print(result[0]["generated_text"])

# SENTIMENT ANALYSIS
classifier = pipeline("sentiment-analysis")
result = classifier("This AiBytec course is the best investment I've made!")
print(result)  # [{'label': 'POSITIVE', 'score': 0.9998}]

# TRANSLATION — English to Urdu
translator = pipeline("translation", model="Helsinki-NLP/opus-mt-en-ur")
result = translator("Artificial Intelligence is transforming education in Pakistan.")
print(result[0]["translation_text"])

# QUESTION ANSWERING
qa = pipeline("question-answering")
result = qa(question="Where is AiBytec based?",
             context="AiBytec is Pakistan's premier AI education platform based in Karachi.")
print(result["answer"])  # Karachi

from sentence_transformers import SentenceTransformer
from sklearn.metrics.pairwise import cosine_similarity

model = SentenceTransformer("all-MiniLM-L6-v2")

sentences = [
    "How do I learn Python for AI?",
    "What is the best way to start coding in Python?",
    "What is the capital city of Australia?"
]
embeddings = model.encode(sentences)

sim_01 = cosine_similarity([embeddings[0]], [embeddings[1]])[0][0]
sim_02 = cosine_similarity([embeddings[0]], [embeddings[2]])[0][0]
print(f"Similarity (Python questions): {sim_01:.3f}")  # ~0.85
print(f"Similarity (unrelated): {sim_02:.3f}")       # ~0.12

import streamlit as st
from groq import Groq
import os

st.set_page_config(page_title="AiBytec AI App", page_icon="🤖")
st.title("🤖 AI Assistant — AiBytec")

client = Groq(api_key=os.getenv("GROQ_API_KEY"))

question = st.text_area("Ask me anything", height=120)
model    = st.selectbox("Model", ["llama3-70b-8192", "mixtral-8x7b-32768"])
temp     = st.slider("Temperature", 0.0, 1.5, 0.7)

if st.button("✨ Get Answer", type="primary"):
    with st.spinner("AI is thinking..."):
        response = client.chat.completions.create(
            model=model,
            messages=[{"role": "user", "content": question}],
            temperature=temp
        )
        answer = response.choices[0].message.content
    st.success("Done!")
    st.markdown(answer)
    st.download_button("💾 Download Answer", answer, "answer.txt")

# Run with: streamlit run app.py

import gradio as gr
from groq import Groq
import os

client = Groq(api_key=os.getenv("GROQ_API_KEY"))

SYSTEM = """You are AiBytec Assistant — a helpful AI tutor.
You help Pakistani students learn Generative AI and Python.
Be encouraging, give clear explanations, use local examples."""

def chat(user_message, history):
    messages = [{"role": "system", "content": SYSTEM}]
    for user, assistant in history:
        messages.append({"role": "user", "content": user})
        messages.append({"role": "assistant", "content": assistant})
    messages.append({"role": "user", "content": user_message})
    response = client.chat.completions.create(
        model="mixtral-8x7b-32768", messages=messages
    )
    return response.choices[0].message.content

demo = gr.ChatInterface(
    fn=chat, title="🎓 AiBytec AI Tutor",
    examples=[
        "What is the difference between supervised and unsupervised learning?",
        "How do I build my first AI app in Python?",
        "What AI skills should I learn for a job in 2025?"
    ],
    theme=gr.themes.Soft()
)
demo.launch(share=True)  # share=True → public URL!

import faiss, numpy as np
from sentence_transformers import SentenceTransformer
from groq import Groq
import os

embedder   = SentenceTransformer("all-MiniLM-L6-v2")
llm_client = Groq(api_key=os.getenv("GROQ_API_KEY"))

class SimpleRAG:
    def __init__(self):
        self.chunks = []; self.index = None

    def add_text(self, text, source="manual", chunk_size=500):
        words = text.split()
        for i in range(0, len(words), chunk_size):
            chunk = " ".join(words[i:i+chunk_size])
            if len(chunk.strip()) > 100:
                self.chunks.append({"text": chunk, "source": source})

    def build_index(self):
        texts = [c["text"] for c in self.chunks]
        embs  = embedder.encode(texts)
        self.index = faiss.IndexFlatL2(embs.shape[1])
        self.index.add(embs.astype(np.float32))

    def answer(self, question, top_k=3):
        q_emb = embedder.encode([question]).astype(np.float32)
        _, ids = self.index.search(q_emb, top_k)
        context = "\n\n".join([self.chunks[i]["text"] for i in ids[0]])
        response = llm_client.chat.completions.create(
            model="llama3-70b-8192",
            messages=[{"role": "user",
                       "content": f"Answer using ONLY the context below.\n\nCONTEXT:\n{context}\n\nQUESTION: {question}"}],
            temperature=0.1
        )
        return response.choices[0].message.content

rag = SimpleRAG()
rag.add_text("AiBytec offers Certificate 1 (Gen AI, PKR 15,000), Certificate 2 (Agentic AI, PKR 45,000). Based in Karachi.")
rag.build_index()
print(rag.answer("How much does Certificate 2 cost?"))

# ─── DATA TYPES ─────────────────────────────────────────────
s = "Hello"              # string
s = f"Name: {name}"      # f-string (always use this)
i = 42                   # integer
f = 3.14                 # float
b = True                  # boolean

# ─── LISTS ──────────────────────────────────────────────────
lst = [1, 2, 3]
lst.append(4)            # [1, 2, 3, 4]
lst[0]                    # 1 — first item
lst[-1]                   # 4 — last item
len(lst)                  # 4

# ─── DICTIONARIES ───────────────────────────────────────────
d = {"key": "value", "num": 42}
d["key"]                  # "value"
d["new"] = "added"        # add new key
d.keys() / .values() / .items()

# ─── CONTROL FLOW ───────────────────────────────────────────
if x > 0:   print("positive")
elif x < 0: print("negative")
else:        print("zero")

for item in my_list:       # iterate list
    print(item)
for i in range(10):        # 0 to 9
    print(i)

# ─── FUNCTIONS ──────────────────────────────────────────────
def add(a, b=10):          # default param
    return a + b

# ─── ERROR HANDLING ─────────────────────────────────────────
try:
    result = risky_api_call()
except Exception as e:
    print(f"Error: {e}")
finally:
    cleanup()             # always runs

# ─── LIST COMPREHENSION ─────────────────────────────────────
squares = [x**2 for x in range(10)]
evens   = [x for x in range(20) if x % 2 == 0]