Artificial Intelligence

Python for AI Applications

Prerequisite:
Python Programming – Fundamentals

Suitable For:

• Python developers who want to build AI-powered applications
• Software engineers transitioning into AI application development
• Students who want practical experience with modern AI APIs
• Entrepreneurs and developers building AI products
• Engineers who want to integrate foundation models into existing software

Description:
Learn how to build modern AI-powered applications with Python using APIs, SDKs, and foundation models. This practical course focuses on integrating existing AI capabilities into real software rather than training models from scratch.

Students Will Learn:

• Connect Python applications to modern AI services
• Work with REST APIs, SDKs, and authentication
• Use leading AI model APIs such as OpenAI and other providers
• Send and process text, image, and structured data
• Generate reliable structured outputs
• Implement function and tool calling
• Manage prompts and model parameters programmatically
• Handle API errors, rate limits, and asynchronous requests
• Build reusable AI application components
• Design practical end-to-end AI applications

Syllabus:

• Introduction to AI application development
• Foundation models and AI APIs
• REST APIs and HTTP fundamentals
• JSON and structured data
• API authentication and secure API-key management
• Python requests and API clients
• Introduction to OpenAI API
• Working with other AI model providers
• Model selection and configuration
• Sending prompts programmatically
• System, user, and application instructions
• Structured outputs and JSON schemas
• Function calling and tool integration
• Working with text and image inputs
• Conversation and context management
• Streaming responses
• Asynchronous API calls
• Error handling, retries, and rate limits
• Token usage, latency, and API cost considerations
• Building reusable AI services
• Practical AI application projects
• Final AI application project

Prompt Engineering for LLMs

Prerequisite:
Basic familiarity with AI applications recommended
Basic Python recommended for hands-on exercises

Suitable For:

• Developers working with LLM-powered applications
• AI engineers
• Product managers and technical professionals using generative AI
• Professionals automating knowledge and content workflows
• Students preparing for LLM and Agentic AI development

Description:
Learn how to communicate effectively with large language models and systematically design, test, and improve prompts for reliable real-world applications. Move beyond trial-and-error prompting toward structured, measurable prompt engineering.

Students Will Learn:

• Understand how LLM behavior affects prompt design
• Write clear and effective system and user instructions
• Apply zero-shot and few-shot prompting techniques
• Control format, context, constraints, and model behavior
• Generate structured and machine-readable outputs
• Design prompts for reasoning and multi-step tasks
• Build reusable prompt templates
• Evaluate prompt quality systematically
• Reduce common LLM failure modes
• Design prompts for production AI applications

Syllabus:

• Introduction to large language models
• How LLMs process instructions and context
• Anatomy of an effective prompt
• System, user, and contextual instructions
• Zero-shot prompting
• Few-shot prompting
• Role and persona prompting
• Context engineering fundamentals
• Prompt decomposition
• Reasoning-oriented prompting techniques
• Structured outputs
• Extraction, classification, summarization, and generation
• Prompt templates and reusable components
• Prompting with external context
• Tool and function-calling prompts
• Managing ambiguity and constraints
• Hallucinations and common failure modes
• Prompt injection and security fundamentals
• Prompt testing and evaluation
• Comparing prompt variants
• Production prompt management
• Hands-on prompt engineering exercises
• Final prompt-engineering project

Generative AI – Foundations & Applications

Prerequisite:
Basic understanding of programming recommended
Python Programming recommended for practical projects

Suitable For:

• Software and AI engineers entering generative AI
• Students who want to understand modern generative models
• Developers who want knowledge beyond simply calling an LLM API
• Technical professionals evaluating generative AI technologies
• Product builders working with text, image, audio, or code generation

Description:
Develop a practical understanding of modern generative AI—from foundation models and transformers to text, image, audio, and code generation. Learn how generative models work conceptually, where they are useful, their limitations, and how they are applied in real-world systems.

Students Will Learn:

• Understand the foundations of generative AI
• Distinguish generative models from traditional predictive AI
• Understand transformers and foundation models at a practical level
• Understand LLMs and modern text-generation systems
• Explore image, audio, and code generation
• Compare major families of generative models
• Understand embeddings and representation spaces
• Select appropriate generative models for different applications
• Understand fine-tuning, RAG, and prompting as different adaptation strategies
• Evaluate quality, limitations, safety, and cost
• Design practical generative AI solutions

Syllabus:

• What is Generative AI?
• Generative AI vs. traditional machine learning
• Evolution of generative models
• Foundation models
• Neural representations and embeddings
• Transformer architecture fundamentals
• Attention and self-attention
• Large Language Models
• Tokens, context windows, and generation
• Text generation
• Code generation
• Image generation
• Diffusion models fundamentals
• Audio and speech generation
• Introduction to multimodal generation
• Prompting as model interaction
• Retrieval-Augmented Generation overview
• Fine-tuning and parameter-efficient adaptation overview
• Model selection: proprietary vs. open models
• Running models through APIs and locally
• Evaluation of generative AI systems
• Hallucinations and reliability
• Responsible AI, copyright, privacy, and safety considerations
• Real-world generative AI use cases
• Practical generative AI projects
• Final application project

Multimodal AI – Vision, Language & Beyond

Prerequisite:

• Python Programming – Fundamentals
• Basic understanding of AI concepts
• Python for AI Applications recommended

Suitable For:

• AI and computer vision engineers
• Developers working with images, text, audio, or video
• Engineers building intelligent multimodal applications
• Robotics developers working with perception and language
• Students interested in modern vision-language systems and foundation models

Description:
Learn how modern AI systems understand and combine multiple forms of information—including text, images, audio, video, and sensor data. Explore multimodal foundation models and build applications that move beyond text-only AI toward systems capable of perceiving and reasoning about the real world.

Students Will Learn:

• Understand multimodal learning and representation
• Work with combinations of text, images, audio, and video
• Understand the foundations of vision-language models
• Use multimodal foundation models through APIs and open models
• Understand image-text embeddings and alignment
• Build visual question-answering applications
• Perform image understanding and structured extraction
• Explore visual grounding and multimodal reasoning
• Understand multimodal RAG
• Connect multimodal perception with intelligent applications
• Understand the role of multimodal AI in robotics
• Build practical multimodal AI projects

Syllabus:

• Introduction to multimodal AI
• Modalities: text, image, audio, video, and sensor data
• Why multimodal learning matters
• Multimodal representations and embeddings
• Transformer fundamentals for multimodal systems
• Vision Transformers (ViT)
• Image-text representation learning
• CLIP and contrastive learning
• Vision-Language Models (VLMs)
• Multimodal foundation models
• Using multimodal AI APIs
• Working with open multimodal models
• Image understanding and description
• Visual question answering
• Document and diagram understanding
• Object grounding and localization
• Multimodal reasoning
• Audio and speech integration
• Video understanding fundamentals
• Multimodal RAG
• Combining visual and textual knowledge
• Multimodal agents
• Introduction to cameras and visual sensor inputs
• Introduction to multimodal AI for robotics
• Model evaluation and failure modes
• Practical multimodal projects
• Final multimodal AI application project

Agentic AI – Project-Based

Focus: RAG, LangChain & LangGraph

Prerequisite:

• Python Programming – Fundamentals
• Python for AI Applications recommended
• Prompt Engineering for LLMs recommended

Suitable For:

• AI and software engineers building autonomous AI applications
• Developers moving from simple LLM applications to agentic systems
• Engineers interested in RAG and knowledge-based AI systems
• Developers building intelligent workflow automation
• Technical professionals exploring multi-agent architectures

Description:
Build intelligent AI systems that can reason over tasks, use tools, retrieve knowledge, maintain state, and coordinate multi-step workflows. This project-driven course progresses from basic LLM applications to production-oriented agentic architectures using RAG, LangChain, and LangGraph.

Students Will Learn:

• Understand AI agents and agentic architectures
• Design goal-oriented AI workflows
• Connect LLMs to external tools and APIs
• Build Retrieval-Augmented Generation systems
• Work with embeddings and vector databases
• Build applications with LangChain
• Design stateful agent workflows with LangGraph
• Implement memory and state management
• Design human-in-the-loop workflows
• Build and coordinate multiple agents
• Evaluate and debug agent behavior
• Build complete agentic AI applications

Syllabus:

• From LLM applications to AI agents
• What makes a system agentic?
• Agent architecture and design patterns
• Models, tools, state, memory, and environment
• Tool calling and external APIs
• Retrieval-Augmented Generation (RAG)
• Document loading and processing
• Chunking and retrieval strategies
• Embeddings
• Vector databases
• Semantic search
• Building RAG pipelines
• RAG evaluation and improvement
• Introduction to LangChain
• Chains, tools, retrievers, and agents
• Introduction to LangGraph
• Nodes, edges, state, and graph-based workflows
• Conditional routing
• Cycles and iterative reasoning
• Memory and persistence
• Human-in-the-loop systems
• Planning and task decomposition
• Multi-agent architectures
• Agent communication and orchestration
• Error handling and guardrails
• Observability, tracing, and evaluation
• Cost and latency considerations
• Project: Knowledge/RAG Agent
• Project: Tool-Using AI Agent
• Project: Multi-Step LangGraph Application
• Final end-to-end Agentic AI project