Back close

Course Detail

Course Name Generative AI With Python
Course Code 26SN601
Program M. Tech. in Cyber Security Systems & Networks
Semester 1
Credits 4
Campus Amritapuri

Syllabus

Syllabus

Python Foundations – Installing Python environment, IDE/editor setup, Variables, data types (int, float, string, boolean), Basic input/output operations, Control flow statements – if/elif/else, for loop, while loop, break/continue/pass, Functions – definition, return statement, scope, variable arguments (*args, **kwargs), Object Oriented Programming – classes, objects, constructors, methods, self keyword, inheritance, polymorphism, encapsulation, abstraction, Intermediate Python concepts – list comprehensions, generators, decorators, context managers, mutable and immutable objects.

Professional Tooling and Engineering Practices – Project setup using Python package managers and virtual environments, Dependency management, Project structure, Logging concepts – logging levels, handlers, formatters, Debugging techniques using IDE tools and pdb, Exception handling – try/except/finally, custom exceptions, Software testing using pytest, Writing unit tests, Running automated test cases, AI-assisted coding workflows, Prompting for code generation, Reviewing and validating AI-generated code.

LLMs – Concepts, APIs and Building with Confidence – Introduction to Large Language Models, Transformer architecture intuition, Pre-training and fine-tuning concepts, Tokenisation, context window, temperature, top-p sampling, hallucination in LLMs, LLM providers and API ecosystems, API authentication and rate limits, Multi-provider integration frameworks, First API integration using Python, Prompt engineering – zero-shot, few-shot, chain-of-thought prompting, System prompts and user prompts, Structured JSON output generation, Debugging with LLM assistance, Observability in LLM applications, Logging prompts, responses, token usage and latency.

LLM Security and Safe AI Development – Prompt injection attacks, jailbreak attacks, indirect prompt injection, Risks in LLM-based applications, Input sanitisation, Output validation, Instruction hierarchy, Secure system prompt design, Secure code generation practices, Risks of executing AI-generated code, OWASP-aware prompting, API key management, Secrets handling in prompts, Supply chain risks in AI frameworks, Building security checklists for LLM applications, Threat modelling for AI systems.

Agentic Applications and Capstone – Introduction to AI agents, Agent loop architecture, ReAct pattern, Tools vs agents vs workflows, Tool use and function calling, Structured tool outputs, Chaining tool calls, Building autonomous agents using Python, Agentic security principles, Principle of least privilege, Safe stopping conditions, Permission checks, Tool output validation, Evaluation and benchmarking of agents – accuracy, latency, cost, LLM-as-judge framework, Building evaluation harnesses, Design and development of capstone project based on agentic AI applications.

Text Books / References
  • Stuart Russell and Peter Norvig, Artificial Intelligence: A Modern Approach, Pearson Education, 4th Edition, 2021.
  • Elaine Rich and Kevin Knight, Artificial Intelligence, McGraw-Hill Education, 3rd Edition.
  • Patrick Henry Winston, Artificial Intelligence, Addison-Wesley, 3rd Edition.
  • Ian Goodfellow, Yoshua Bengio and Aaron Courville, Deep Learning, MIT Press, 2016.
  • Tom M. Mitchell, Machine Learning, McGraw-Hill Education, International Edition.

Introduction

This course equips MTech Cybersecurity systems and networks students with the skills to program in Python from first principles, understand and work with modern Large Language Models (LLMs), and build production-aware, security-conscious generative AI applications. Students will progress from zero Python knowledge to confidently building, evaluating, and securing agentic AI systems capabilities that are now fundamental to the modern software landscape.

Objectives and Outcomes

Course Objectives

  • Write clean, idiomatic Python and structure projects professionally
  • Understand how LLMs work and interact with them via APIs from multiple providers
  • Apply prompt engineering, observability, and debugging practices in AI applications
  • Design and build agentic systems that use tools and perform multi-step reasoning
  • Identify and mitigate security vulnerabilities specific to LLM-powered applications
  • Use modern AI-assisted coding tools (Cursor, OpenCode) effectively and critically

Course Outcomes

CO Description
CO1 Write clean, idiomatic Python using core and intermediate features including OOP, generators, decorators, and context managers.
CO2 Set up and manage Python projects using modern tooling (uv, virtual environments) and apply software engineering practices — logging, debugging, exception handling, and testing.
CO3 Understand how LLMs work (tokenisation, context window, hallucination) and interact with OpenAI, Claude, and Gemini APIs using effective prompt engineering and observability practices.
CO4 Design and build agentic AI applications with tool use, function calling, and multi-step reasoning; evaluate and benchmark their performance.
CO5 Identify and mitigate security vulnerabilities in LLM-powered applications including prompt injection, jailbreaks, insecure code generation, and unsafe agentic behaviours.

Evaluation Pattern

CO-PO Mapping 

Correlation Levels: 3 = High, 2 = Moderate, 1 = Low

COs POs PO1 PO2 PO3
CO1 2 1 3
CO2 2 2 3
CO3 2 2 3
CO4 3 2 3
CO5 3 2 3

Evaluation Pattern – 70:30

  • Midterm Exam – 30% – Hands-On
  • Quizzes – 20%
  • Lab Assignments & Case Study – 10%
  • Certification Course -10%
  • Project – 30%

DISCLAIMER: The appearance of external links on this web site does not constitute endorsement by the School of Biotechnology/Amrita Vishwa Vidyapeetham or the information, products or services contained therein. For other than authorized activities, the Amrita Vishwa Vidyapeetham does not exercise any editorial control over the information you may find at these locations. These links are provided consistent with the stated purpose of this web site.

Admissions Apply Now