AI Agents & Gen AI Course
Three courses in one. You learn by building real AI agents — every concept comes up inside a build, exactly when you need it.
Taught agent-wise and project-wise, not module-wise
The module lists on this page are for reference. They are not covered in the given sequence.
- Modules are not followed chapter by chapter
- Topics appear when a real AI agent project needs them
- Some advanced concepts may come before basic ones, if the project demands it
- You learn how AI agents actually work in real scenarios
- Clear understanding of where and why each concept is used
- Faster transition from learning → building → deploying
- No theoretical overload — only practical, outcome-focused learning
The agents you will build
Some of these are built with you in live classes; some come as recorded builds in your library, covering different concepts. This is the base list — more agents may be added over time.
Customer & sales
Agents that talk to your customers and keep your leads moving.
An agent that replies on WhatsApp by itself, understands what the customer is asking for, and sends the right response.
Where it is used: 24x7 customer support on WhatsApp.
Setting up the WhatsApp account takes time, and Meta’s verification can take days — so this build is given as a recording, with a PDF guide for the verification.
An agent that talks to people on a phone call in a natural voice, and can be used in many different situations.
Where it is used: Customer calls handled by an AI voice.
An agent that reads incoming emails, replies to the routine ones on its own, and drafts the important ones for a person to approve before they go out.
Where it is used: Routine queries handled automatically; your team only looks at the emails that matter.
An agent that keeps your list of leads up to date and follows up with them at the right time.
Where it is used: Never miss a follow-up with a potential customer.
An agent that adds new leads into Zoho CRM automatically, so nothing has to be typed in by hand.
Where it is used: New leads go straight into your CRM.
Content & social media
Agents that write and publish for you, on a schedule.
An agent that writes social media posts on trending topics, or on any topic you give it.
Where it is used: A content engine for your social media.
An agent that writes LinkedIn posts and publishes them for you.
Where it is used: Stay active on LinkedIn without posting by hand.
An agent that creates posts and publishes them on your Facebook page automatically.
Where it is used: A Facebook page that keeps posting on its own.
The Facebook account setup differs from person to person and takes time, so this build is given as a recording.
An agent that creates posts and publishes them on your Instagram account automatically.
Where it is used: An Instagram account that keeps posting on its own.
The Instagram account setup differs from person to person and takes time, so this build is given as a recording.
After you have edited a video, this agent writes the description and hashtags and uploads the video to YouTube for you.
Where it is used: Everything after editing, done automatically.
Knowledge & data
Agents that answer from your own files and numbers.
An agent that reads your own documents and answers questions from them — so it replies from your files, not from guesswork.
Where it is used: A helpdesk chatbot for a company, with access to its internal documents.
An agent that reads a data file, such as a spreadsheet, and gives you insights and suggestions from it.
Where it is used: Data analysis for individuals and small businesses, without hiring an analyst.
Everyday automation
Small agents that quietly take a job off your plate.
When someone fills a form, this agent sends them the right document by email, automatically.
Where it is used: Documents reach the right people the moment a form is submitted.
An agent that makes a branded greeting image for a birthday or an event and sends it out on its own.
Where it is used: Personal greetings for your customers, or your own circle, without doing it by hand.
What you learn while building these agents
These are the building blocks that come up inside the agent builds above — taught when an agent needs them, not as separate chapters. Everything here is on the n8n platform; much of it applies to other platforms too.
- Agent logic — The difference between a simple automation and an agent that can reason and decide.
- Workspace setup — Getting comfortable with the n8n interface and setting up your environment.
- How data flows — How data (JSON) moves from one step to the next and gets processed on the way.
- Authentications — Connecting safely to outside tools — LLMs, Gmail, WhatsApp, Telegram, CRMs — with OAuth2, API keys and tokens.
- Manual & chat trigger — Start an agent by hand, or from a chat message.
- Webhook trigger — Start an agent the moment data arrives from any outside source.
- Schedule trigger — Run an agent automatically at set times.
- n8n form trigger — Start an agent from a simple form you build in minutes.
- App triggers — Start an agent when something happens inside an app you use — the 5–6 most used ones.
- The AI Agent node — The central node that brings together tools, memory and the LLM.
- Working with LLMs — Connecting an LLM to your agent and tuning it for consistent, reliable answers.
- Embeddings — Preparing your own data so the agent can search through it.
- Other models — Using other and open-source AI models with n8n.
- Simple memory — Short-term memory, so follow-up messages make sense.
- Vector store memory (RAG) — Long-term memory — a knowledge base built from your own data.
- Set node — Changing and shaping data with n8n expressions.
- IF / Switch — Decision-making, so the agent takes the right path.
- HTTP Request — Talking to any outside service through its API — the most important tool of all.
- Code node — A little custom logic for the cases a standard node cannot handle.
- Split & Merge — Sending output to many steps at once, or joining results back together.
- Error handling — What the agent does when a step fails.
- Merging parallel data — Combining results from steps that ran side by side.
- Wait & human-in-the-loop — Pausing for a person’s approval, or for a set time.
- Sub-workflows — Breaking big agents into reusable parts.
- Security & best practices — Keeping credentials safe and documenting workflows properly.
Platforms covered for AI agents
n8n is the base. The others are taught so that your skill never depends on a single platform.
Reduces platform dependency.
Practical, hands-on approach.
Each platform taught by a specialist.
Industry-ready tools.
Prompt Engineering for AI Agents
Nine sections, from what a prompt really is to defending an agent against prompt attacks.
Learn what prompting really means, how your prompt influences the output, and why clear instructions really matter.
Outcome: Build a strong base understanding of prompting and a clarity mindset.
Shift from merely “asking questions” to “designing conversations” with intent, context, and goals.
Outcome: Start thinking like a prompt engineer, and not just a casual user.
Break down prompts into instruction, context, examples, and output framing — the essential building blocks of a prompt.
Outcome: Stop writing random prompts, and start building prompts that are structured.
Use reusable prompt frameworks to standardise your approach across tasks and industries.
Outcome: Create professional, repeatable prompts quickly.
Master Zero-Shot, Few-Shot, Chain-of-Thought, ReAct, Self-Consistency, and more prompting techniques.
Outcome: Handle complex reasoning and advanced tasks with ease using prompting techniques.
Learn proven patterns like Persona, Template, Recipe, Outline, Context Layering and more.
Outcome: Understand and implement the prompt patterns in your tools and LLMs.
Break complex tasks into multi-step prompts that deliver higher accuracy and deeper reasoning.
Outcome: Learn to create prompts by breaking complex problems into multiple steps.
Use variables, dynamic templates, and visual prompts to enhance output quality.
Outcome: Design prompts that adapt to users, inputs, images, and contexts.
Understand prompting attacks on AI agents, like prompt injection, jailbreaking etc., and how to prevent them.
Outcome: Learners are introduced to agent vulnerabilities, and how to reduce the risk of prompting attacks on an AI agent.
Gen AI tools — what they do, and where they are used
One live class every week is a Gen AI tool class. These are the areas covered.
The ultimate creative suite to generate professional images and graphic designs from text prompts.
Industries / areas: Graphic design, digital marketing, content creation, film making, social media, advertising.
Midjourney and moreCreate stunning and realistic videos completely using AI, through your prompts.
Industries / areas: Graphic design, digital marketing, content creation, film making, social media, advertising.
KlingAI · Flow · RunwayAutomate your office tasks, manage emails, and analyse complex business data from spreadsheets just by asking plain-English questions.
Industries / areas: Data analytics, business operations, finance, marketing, sales — all corporate roles.
Julius · Excel · CopilotInstantly generate high-quality music in any genre, complete with vocals, just from a text description, lyrics, and the style.
Industries / areas: Music production, film making (soundtracks), content creation (YouTube), game development, advertising.
Suno · UdioCreate professional, studio-quality videos featuring realistic AI avatars speaking your script — for training or marketing.
Industries / areas: Corporate training, learning & development, marketing, YouTube content creation, news and media.
HeyGen · D-IDCreate hyper-realistic voice-overs, clone your own voice, or edit entire podcasts as easily as editing a text document.
Industries / areas: Podcasting, e-learning, voice-over & dubbing, digital marketing, content creation, news and media, advertising.
ElevenLabs · HumeA universal toolkit to research topics, write content, draft clinical notes, and create apps using AI.
Industries / areas: Education, content creation, marketing, project students — applicable to multiple industries.
Advanced LLM tools for writing, research, coding assistance, and decision-making with high accuracy and speed.
Industries / areas: Business, education & training, marketing, software development, operations, entrepreneurship, content creation.
ChatGPT · Grok · Copilot · Claude · GeminiCreate presentations, pitch decks and documents, with visual content, in minutes with AI.
Industries / areas: IT industry, business, education, digital marketing, social media, branding, content creation.
Gamma · Google Slides · CanvaAI-powered productivity platforms to organise knowledge, manage projects and streamline team collaboration.
Industries / areas: Project management, operations, startups, agencies, remote teams, documentation, knowledge management, product development.
Notion · ClickUpAll product names and trademarks belong to their respective owners. GyaniPandit teaches these tools and is not affiliated with, partnered with or endorsed by any of them.
Select your preferred batch
Live classes cover the important topics; everything else is in your recorded library. Doubt sessions are open to both batches.
9:00 PM – 10:30 PM
2 PM – 7 PM
11:30 AM – 1 PM
2 PM – 7 PM
Three courses, three certificates
Each certificate carries its own credential ID and can be verified on our site at any time.
Questions about the syllabus or the batches? Call or message us.
Verify a certificate · © 2026 GyaniPandit · gyanipandit.com