Predictive model lab
End-to-end ML project from EDA to deployed pickle or API endpoint.
Intermediate
AI & Machine Learning
You learn when ML helps and when a simple rule or API call is enough. We cover data prep, classical models, intro deep learning, and practical LLM integration so you can ship assistants and search tools on top of real business data.
- Duration
- 4 months
- Batch size
- 8 to 12 students per batch
What you will learn
- Prepare and explore datasets with pandas
- Train and evaluate classification and regression models
- Understand neural network basics with a small project
- Integrate OpenAI or open-source LLM APIs safely
- Build a RAG or chatbot capstone over approved sample data
Course modules
Data science with Python
4 weeks- NumPy and pandas workflows
- Visualization with matplotlib
- Feature engineering basics
- Train/test splits and metrics
Machine learning core
4 weeks- Scikit-learn pipelines
- Classification and regression
- Clustering and dimensionality reduction
- Model selection and overfitting
Deep learning intro
3 weeks- Neural network intuition
- TensorFlow or PyTorch basics
- CNN intro for image tasks
- Transfer learning overview
LLMs and AI applications
4 weeks- Prompt design and evaluation
- Embeddings and vector search
- RAG architecture patterns
- Capstone AI assistant or search tool
Who it is for
- Python developers entering ML or AI roles
- Product engineers adding AI features to existing apps
- Analysts moving toward data science work
Prerequisites
- Python programming (our Python track or equivalent)
- Basic statistics and algebra comfort
Projects you will build
Document Q&A assistant
RAG pipeline over PDFs with citation-style answers and guardrails.
Tools and stack: Python, Jupyter, scikit-learn, TensorFlow/PyTorch intro, OpenAI API, Chroma or FAISS
Questions about this track?
Tell us your background and preferred batch timing. We will confirm prerequisites and seat availability.