Welcome to the AI & ML Bootcamp by iCodeGuru — a hands-on, project-driven journey from Python fundamentals to cutting-edge GenAI with LangChain, Transformers, and beyond.
This Bootcamp combines concept mastery with real-world application, preparing you for the AI revolution.
Explore each module’s concepts, practice areas, and tools below. Each folder contains relevant notebooks, assignments, and practice files — not listed here for brevity.
Topics Covered:
- Variables, data types, and operators
- Control flow:
if,for,while - Functions, lambda, decorators
- Lists, tuples, sets, dictionaries
- File I/O and error handling
- Classes, objects, and OOP principles
✅ Foundation for everything ahead.
Topics Covered:
- Series vs DataFrames
- Data cleaning & transformation
- GroupBy, aggregation, merging
- Working with real datasets
- Basic exploratory data analysis (EDA)
🔍 From raw CSV to clean, insightful data.
Tools Used:
Matplotlibfor core plottingSeabornfor statistical graphsPlotlyfor interactive charts
Skills Gained:
- Bar, line, pie, scatter plots
- Histograms, heatmaps, correlation plots
- Plot customization and storytelling
🎨 Turn data into clear visual stories.
Topics Covered:
- Array creation and manipulation
- Indexing, slicing, and reshaping
- Broadcasting and vectorization
- Statistical operations and matrix math
🧮 Essential for ML pipelines and performance.
Supervised Models:
- Linear & Logistic Regression
- KNN, Decision Trees, Random Forest
- SVM and Naive Bayes
Unsupervised Models:
- K-Means, Hierarchical Clustering
- DBSCAN, PCA
Hands-On:
- Model evaluation: accuracy, precision, recall, F1
- Confusion matrices, ROC curves
- Sklearn pipelines & hyperparameter tuning
📊 From theory to full model training and testing.
Projects Built:
- CNN for image classification (Fashion MNIST)
- LSTM/RNN for sentiment analysis (IMDb)
- GAN for digit generation (MNIST)
Concepts Learned:
- Activation functions, optimizers
- Dropout, batch normalization
- Model training, validation, and evaluation
⚙️ End-to-end deep learning mastery.
Skills Practiced:
- Using chat models via OpenAI or Groq
- Designing prompt templates
- Creating multi-step chains
- Building RAG (Retrieval-Augmented Generation) apps
- Implementing LLM Agents with memory and tools
🧠 LLM orchestration like a pro.
Learned Techniques:
- Zero-shot, one-shot, few-shot prompts
- Role prompting & instruction tuning
- Prompt templates for classification, reasoning, coding, Q&A
- Prompt testing for performance & consistency
💡 Crafting the perfect instructions for GenAI models.
Hands-On With:
transformersandpipeline()API- Text classification, translation, summarization
- Tokenizers and model configuration
- Model comparisons and use-case matching
🤖 Using state-of-the-art models in minutes.
Explored Concepts:
- Tokenization, stemming, lemmatization
- Stopword removal, POS tagging, NER
- TF-IDF, Bag-of-Words
- Built a mini NLP pipeline using NLTK and Scikit-learn
📚 Before transformers, came the classics.
| Type | Tools & Libraries |
|---|---|
| Core Language | Python |
| Data Analysis | Pandas, NumPy |
| Visualization | Matplotlib, Seaborn, Plotly |
| ML/DL | Scikit-learn, TensorFlow, Keras |
| NLP & GenAI | NLTK, Hugging Face Transformers, LangChain |
| Utilities | Jupyter, VSCode, Git |
Built as part of the iCodeGuru Python & AI Bootcamp
“Learn by building. Master by sharing.”
Designed for developers, data enthusiasts, and aspiring AI engineers.
Share your progress, collaborate, and contribute to the AI community.
Happy Learning 🚀