Register Now
Home / Courses / Generative AI & Agentic AI

Generative AI & Agentic AI Program

A 15-module applied syllabus — from Python for Generative AI and core ML/DL foundations, through LLMs, prompt engineering, RAG, image generation, AI agents and MCP, all the way to fine-tuning, cloud deployment, LLMOps and enterprise AI leadership.

Program Overview

From prompting to production-grade agentic automation.

This program builds job-ready capability in Generative AI and Agentic AI from the ground up: Python for Generative AI, machine learning & deep learning foundations, NLP, and working with LLMs like ChatGPT, before moving into workflow automation (n8n), prompt engineering, LangChain & RAG, image generation with Stable Diffusion, AI agents & multi-agent systems, and the Model Context Protocol (MCP).

The program then goes further than most Generative AI courses — covering Small Language Models (SLMs), LLM fine-tuning, an overview of cloud platforms for Generative AI, LLMOps, and Enterprise AI Leadership — so learners finish not just able to build with Generative AI, but able to fine-tune, deploy, operate and champion it at an organizational level. Best suited for learners with basic Python familiarity targeting roles such as AI Engineer, Agentic AI Developer, Generative AI Solutions Architect, or LLMOps Engineer.

🐍 Python for Generative AI 💬 LLMs & ChatGPT ⚙️ n8n Automation 🔗 LangChain & RAG 🎨 Stable Diffusion 🤝 AI Agents & MCP 🧬 SLMs & Fine-Tuning ☁️ Cloud & LLMOps
ModeLive Online
LevelIntermediate · AI/ML Learners Progressing to Generative AI
Duration~13 Weeks (~88 Contact Hrs)
Modules15
PrerequisitesAI/ML foundations + basic Python recommended
CertificateDataSpeaks Certificate
FeeContact for current pricing
Register Interest
How the Program Is Built

Structure, outcomes & assessment at a glance

The full week-by-week, topic-by-topic curriculum unlocks after registration below — here's the shape of the program.

🗺️

Program Structure

  • Modules 1–4: Foundations — Python for Generative AI, ML & Deep Learning, NLP, LLMs & ChatGPT
  • Modules 5–8: Building with Generative AI — n8n automation, prompt engineering, LangChain & RAG, Stable Diffusion
  • Modules 9–11: Agentic AI — AI agents & multi-agent systems, MCP, Small Language Models
  • Modules 12–14: Production-readiness — LLM fine-tuning, cloud platforms, LLMOps
  • Module 15: Enterprise AI Leadership — governance, ROI & adoption strategy
🎯

Learning Outcomes

  • Write Python for Generative AI workflows and explain the ML/DL & NLP foundations behind LLMs
  • Use LLMs/ChatGPT, prompt engineering and LangChain/RAG to build grounded AI applications
  • Generate images with Stable Diffusion and build single- and multi-agent systems with MCP tool access
  • Work with Small Language Models and fine-tune an LLM on a custom dataset
  • Deploy and operate Generative AI apps on the cloud with LLMOps practices, and frame enterprise AI adoption & ROI
📝

Assessment

  • Module-end hands-on labs (formative assessment)
  • Practical evaluations for agent-building, fine-tuning & deployment modules
  • Final Generative AI/Agentic AI-focused mock interview with structured feedback
Full Curriculum

15 modules, week by week — unlocked after registration

Register for this course to unlock the complete module-by-module syllabus, learning objectives, tools and reference documents.

🔒

Full Curriculum Locked

Register your interest in the Generative AI & Agentic AI program to unlock the complete 15-module syllabus — learning objectives, detailed topics, tools and module outcomes for every week.

Register to Unlock
1

Python for Generative AI

Week 1
6 hrs
Learning Objectives
  • Write Python code confidently for AI/Generative AI workflows
  • Set up and use API-key based access to Generative AI model providers securely
  • Build small utility scripts that prepare data/prompts for downstream Generative AI tasks
Detailed Topics
  • Python refresher: functions, classes, virtual environments, package management
  • Working with JSON, environment variables, and API keys/secrets safely
  • Core libraries for Generative AI work: requests, NumPy, Pandas for data prep
  • Calling your first LLM API (OpenAI/Anthropic/Gemini) from a Python script
  • Async and streaming basics in Python for handling LLM responses
  • Setting up a dev environment: VS Code, Jupyter/Colab, virtual environments
Tools & Technologies
Python 3 Jupyter / Google Colab VS Code pip / conda
Module Outcome: Learners can set up a Python environment from scratch and write a script that securely calls an LLM API and processes the JSON response.
2

Machine Learning and Deep Learning

Week 1–2
8 hrs
Learning Objectives
  • Understand core ML concepts as the foundation for Generative AI
  • Understand neural network fundamentals: layers, activations, backpropagation
  • Recognize how deep learning underpins modern generative models
Detailed Topics
  • ML recap: regression, classification, train/test split, evaluation metrics
  • Neural network basics: perceptrons, multi-layer networks, activation functions
  • Backpropagation and gradient descent, explained intuitively
  • Introduction to CNNs and RNNs, and why transformers replaced RNNs for language
  • Overfitting, regularization, and why data quality matters for Generative AI models
  • Hands-on: train a simple neural network on a sample dataset
Tools & Technologies
scikit-learn TensorFlow / Keras or PyTorch Google Colab
Module Outcome: Learners can explain how a neural network learns and train a basic model end-to-end on a sample dataset.
3

NLP (Natural Language Processing)

Week 2–3
6 hrs
Learning Objectives
  • Apply core NLP preprocessing techniques to real text data
  • Understand how text is represented numerically for machine learning
  • Build a basic NLP pipeline for a real task
Detailed Topics
  • Text preprocessing: tokenization, stemming, lemmatization, stopword removal
  • Text representation: Bag-of-Words, TF-IDF, word embeddings (Word2Vec/GloVe)
  • Introduction to sequence models and attention mechanisms
  • Named Entity Recognition, POS tagging, and basic text classification
  • How NLP evolved into today's transformer-based LLMs
  • Hands-on: build a sentiment analysis or text classification pipeline
Tools & Technologies
NLTK / spaCy scikit-learn Python
Module Outcome: Learners can preprocess raw text and build a working NLP classification pipeline with measurable accuracy.
4

LLM and ChatGPT

Week 3
6 hrs
Learning Objectives
  • Understand how large language models like ChatGPT are built and function
  • Use the ChatGPT/OpenAI API effectively for real tasks
  • Understand key LLM parameters and their effect on output
Detailed Topics
  • Transformer architecture explained conceptually: attention, tokens, context window
  • How ChatGPT/GPT models are trained: pretraining, fine-tuning, RLHF
  • Using the OpenAI API: chat completions, system/user/assistant roles
  • Key parameters: temperature, top-p, max tokens, frequency/presence penalty
  • Practical use cases: summarization, Q&A, content generation, coding help
  • Limitations: hallucination, context limits, knowledge cutoffs
Tools & Technologies
OpenAI API ChatGPT Python Postman
Module Outcome: Learners can integrate the ChatGPT/OpenAI API into a Python script and tune parameters to get reliable outputs for a defined task.
5

n8n — AI Workflow Automation

Week 4
5 hrs
Learning Objectives
  • Understand workflow automation concepts and where n8n fits
  • Build automated workflows connecting apps, APIs, and AI models
  • Design trigger-based automations for real business use cases
Detailed Topics
  • Introduction to n8n: nodes, triggers, workflows, executions
  • Connecting APIs and apps (Slack, email, Sheets) into a single workflow
  • Adding AI/LLM nodes into an n8n workflow for automated processing
  • Conditional logic, branching, and error handling in workflows
  • Scheduling and webhook-triggered automations
  • Hands-on: build an end-to-end automated workflow with an AI step
Tools & Technologies
n8n Webhooks REST APIs
Module Outcome: Learners can design and deploy a working n8n workflow that automates a multi-step task involving at least one AI/LLM node.
6

Prompt Engineering

Week 4–5
5 hrs
Learning Objectives
  • Construct clear, reliable prompts for varied business and technical tasks
  • Apply advanced prompting techniques to improve output quality and consistency
  • Recognize and mitigate common prompt failure modes
Detailed Topics
  • Anatomy of a strong prompt: role, context, task, constraints, output format
  • Zero-shot, few-shot, and chain-of-thought prompting
  • System prompts vs user prompts, and reusable prompt templates
  • Structured output prompting: requesting JSON for automation
  • Prompt chaining: breaking a complex task into smaller prompts
  • Common failure modes: hallucination, ambiguity, prompt injection awareness
Tools & Technologies
ChatGPT / Claude / Gemini PromptLayer (optional)
Module Outcome: Learners build a reusable prompt library covering at least 3 distinct business tasks, each with a tested, refined template.
7

LangChain and RAG

Week 5–6
7 hrs
Learning Objectives
  • Build LLM-powered applications using LangChain
  • Design and implement a Retrieval-Augmented Generation (RAG) pipeline
  • Understand chunking, embeddings, and vector retrieval for grounding LLMs
Detailed Topics
  • LangChain core concepts: chains, prompts, memory, tools
  • Why RAG: reducing hallucination and grounding responses in real documents
  • Document ingestion, chunking strategies, and embedding generation
  • Vector stores (Chroma/FAISS) and semantic retrieval
  • Building a RAG chatbot that answers from custom documents
  • Evaluating RAG output quality and tuning retrieval parameters
Tools & Technologies
LangChain ChromaDB / FAISS Embedding models Python
Module Outcome: Learners build and demo a working RAG chatbot that correctly answers questions grounded in a provided document set.
8

Large Vision Models and Stable Diffusion

Week 6–7
6 hrs
Learning Objectives
  • Understand how large vision and image-generation models work conceptually
  • Use Stable Diffusion to generate and edit images from text prompts
  • Apply prompt engineering techniques specific to image generation
Detailed Topics
  • Overview of large vision models (CLIP, Vision Transformers) and multimodal AI
  • How diffusion models generate images: noise, denoising steps, latent space
  • Using Stable Diffusion: text-to-image, image-to-image, inpainting
  • Prompting for images: style, composition, negative prompts
  • Customization basics: LoRA and textual inversion (conceptual overview)
  • Ethical and copyright considerations in AI-generated imagery
Tools & Technologies
Stable Diffusion Hugging Face
Module Outcome: Learners can generate and refine AI images using Stable Diffusion for a defined creative or business use case.
9

AI Agent and Multi-AI Agents

Week 7–8
7 hrs
Learning Objectives
  • Understand what makes an AI system 'agentic' and its core components
  • Build a single tool-using AI agent
  • Design a multi-agent system where agents collaborate on a task
Detailed Topics
  • Agent architecture: planning, memory, tool use, the ReAct reasoning loop
  • Building a single agent with tool access using LangChain
  • Multi-agent orchestration: roles, delegation, and collaboration patterns
  • Framework overview: CrewAI and AutoGen for multi-agent systems
  • Common pitfalls: infinite loops, redundant work, runaway API cost
  • Hands-on: build a multi-agent system (e.g., researcher + writer + reviewer)
Tools & Technologies
LangChain CrewAI AutoGen Python
Module Outcome: Learners deliver a working multi-agent system where 2–3 agents collaborate with clear role separation to complete a task.
10

MCP (Model Context Protocol)

Week 8–9
5 hrs
Learning Objectives
  • Understand the Model Context Protocol (MCP) and why it exists
  • Connect an AI agent to external tools/data sources using MCP
  • Understand security considerations when exposing tools to AI agents
Detailed Topics
  • What MCP is and how it standardizes agent-to-tool/data connections
  • MCP servers and clients: how an agent discovers and calls tools via MCP
  • Comparing MCP to traditional function calling/tool calling
  • Setting up and connecting to a sample MCP server
  • Building a simple custom MCP tool/server
  • Security: scoping permissions, validating inputs, safe tool exposure
Tools & Technologies
MCP servers/SDKs Python Claude / MCP-compatible clients
Module Outcome: Learners connect an AI agent to at least one MCP server and demonstrate a successful tool call through the protocol.
11

SLM (Small Language Models)

Week 9
5 hrs
Learning Objectives
  • Understand what Small Language Models are and when to use them over LLMs
  • Evaluate tradeoffs between SLMs and large models (cost, latency, privacy)
  • Deploy and run a small language model for a defined task
Detailed Topics
  • What defines an SLM: parameter count, use cases, example model families
  • Why SLMs matter: on-device AI, lower cost, faster inference, data privacy
  • Running an SLM locally or on modest hardware
  • Comparing SLM vs LLM output quality for specific tasks
  • Use cases: edge deployment, offline assistants, cost-sensitive applications
  • Hands-on: run and evaluate a small language model on a sample task
Tools & Technologies
Hugging Face Local inference tools (e.g. Ollama) Python
Module Outcome: Learners can deploy a small language model locally and evaluate it against an LLM for a defined use case.
12

LLM Fine-Tuning

Week 9–10
6 hrs
Learning Objectives
  • Understand when fine-tuning is (and isn't) the right approach vs prompting/RAG
  • Understand the fine-tuning workflow at a practical level
  • Fine-tune a model on a custom dataset for a specific task
Detailed Topics
  • Fine-tuning vs prompting vs RAG: choosing the right approach
  • Preparing a fine-tuning dataset: format, size, quality considerations
  • Parameter-efficient fine-tuning: LoRA and QLoRA (conceptual + hands-on)
  • Fine-tuning workflow: base model selection, training, evaluation
  • Common pitfalls: overfitting, catastrophic forgetting, cost management
  • Hands-on: fine-tune a small/open model on a custom dataset
Tools & Technologies
Hugging Face Transformers LoRA / PEFT Google Colab
Module Outcome: Learners fine-tune a model on a custom dataset and demonstrate measurable improvement on the target task.
13

Overview of Cloud Platforms for Generative AI

Week 10–11
5 hrs
Learning Objectives
  • Understand Generative AI offerings across major cloud platforms
  • Choose the right cloud service for a given Generative AI use case
  • Deploy a simple Generative AI application on a cloud platform
Detailed Topics
  • Generative AI services overview: Azure OpenAI, AWS Bedrock, Google Vertex AI
  • Managed vs self-hosted model deployment tradeoffs
  • Cost and scaling considerations across cloud providers
  • Basic deployment: hosting an API-backed Generative AI app on the cloud
  • Security and access management for cloud-hosted AI services
  • Comparing providers for a given project's needs
Tools & Technologies
Azure OpenAI AWS Bedrock Google Vertex AI
Module Outcome: Learners can compare major cloud Generative AI platforms and deploy a basic Generative AI application on at least one of them.
14

LLMOps

Week 11–12
6 hrs
Learning Objectives
  • Understand the operational lifecycle of LLM-powered applications in production
  • Apply monitoring, logging, and evaluation practices to Generative AI systems
  • Manage cost, latency, and reliability of deployed LLM applications
Detailed Topics
  • LLMOps vs traditional MLOps: what's different for generative systems
  • Monitoring and logging LLM/agent behavior in production
  • Evaluation pipelines: automated and human-in-the-loop evaluation
  • Cost and latency optimization for production LLM applications
  • Versioning prompts, models, and pipelines
  • CI/CD considerations for Generative AI applications
Tools & Technologies
LangSmith Monitoring / logging tools Cloud dashboards
Module Outcome: Learners can set up basic monitoring and evaluation for a deployed LLM application and identify cost/latency optimization opportunities.
15

Enterprise AI Leadership

Week 12–13
5 hrs
Learning Objectives
  • Understand how enterprises plan and govern AI/Generative AI adoption
  • Apply responsible AI and governance principles at an organizational level
  • Communicate AI strategy and ROI to business stakeholders
Detailed Topics
  • Enterprise AI adoption roadmap: from pilot to scale
  • AI governance: responsible AI, compliance, and data privacy at organizational level
  • Building a business case and measuring ROI for Generative AI initiatives
  • Change management: upskilling teams and driving AI adoption internally
  • Vendor selection and build-vs-buy decisions for enterprise Generative AI
  • Case discussion: real-world enterprise AI adoption patterns
Tools & Technologies
AI governance frameworks Case studies
Module Outcome: Learners can outline an enterprise AI adoption roadmap and articulate governance considerations and ROI framing for a Generative AI initiative.
Outcomes

What you'll walk away with

✅

Skills You'll Gain

  • Solid Python, ML/DL and NLP foundations that support everything built on top in Generative AI
  • Practical LLM & ChatGPT integration, prompt engineering, and LangChain/RAG application-building
  • Image generation with Stable Diffusion, and single/multi-agent systems using MCP tool access
  • Hands-on experience with Small Language Models and LLM fine-tuning (LoRA/QLoRA)
  • Cloud deployment and LLMOps practices, plus enterprise AI governance & ROI framing
🎯

Career Paths This Opens

  • AI Engineer
  • Agentic AI Developer
  • Generative AI Solutions Architect
  • LLMOps / AI Operations Engineer
  • Enterprise AI Consultant

Ready to build your first autonomous AI agent?

Register your interest — our team will share batch dates, structure and fee details.

Register Now Explore AI/ML & Automation Instead →
DataSpeaks Assistant
Hi! How can I help you today? Ask me about courses, fees, certification or registration — I'll answer instantly. For anything else, use the Contact form and our team will follow up.