Generative AI with Claude: Build AI Agents, RAG and MCP Apps
Hands-on labs and a capstone project for developers and QA engineers with basic Python or TypeScript.
Trainer: Kiran Kumar

Registrations Open
Free Demo Sessions
- Mon, Nov 16, 2026 · 6:00 – 7:00 PM PSTTue, Nov 17 · 7:30 – 8:30 AM IST
- Tue, Nov 17, 2026 · 6:00 – 7:00 AM PSTTue, Nov 17 · 7:30 – 8:30 PM IST
- Tue, Nov 17, 2026 · 6:00 – 7:00 PM PSTWed, Nov 18 · 7:30 – 8:30 AM IST
- Wed, Nov 18, 2026 · 6:00 – 7:00 AM PSTWed, Nov 18 · 7:30 – 8:30 PM IST
- Thu, Nov 19, 2026 · 6:00 – 7:00 PM PSTFri, Nov 20 · 7:30 – 8:30 AM IST
- Fri, Nov 20, 2026 · 6:00 – 7:00 AM PSTFri, Nov 20 · 7:30 – 8:30 PM IST
$199 early bird with code GENAI-199 · $299 regular
Register Now- Live instructor-led sessions
- A hands-on lab in every module
- Capstone project with review
- Certificate of completion
Course facts
- Format
- Live online, instructor-led, with hands-on labs
- Level
- Intermediate — basic Python or TypeScript assumed
- Prerequisites
- Basic Python or TypeScript, REST and JSON knowledge, and an Anthropic Console account with API credits
- API credits
- Not included — labs run on your own Anthropic account and are billed by Anthropic
- Schedule
- Free demos on 16, 17, 18, 19 and 20 November 2026; paid program runs 30 November to 18 December 2026
- Price
- $199 early bird with code GENAI-199 at checkout · $299 regular, in USD
- Trainer
- Kiran Kumar
- Certificate
- Certificate of completion after the module labs and capstone presentation
- Included
- All ten live modules, a lab in every module, assignments with expert feedback and capstone review
Prerequisites
Three things before the first lab. No previous AI or machine-learning experience is needed.
Basic Python or TypeScript
You should be comfortable with functions, packages and handling JSON. Every lab is provided in both languages.
REST and JSON knowledge
You should be able to call an HTTP API, set headers and read a JSON response. There is no web framework work in the course.
Anthropic account with API credits
Create a Claude Console account and add credits before the first lab. API usage is billed by Anthropic and is not included in the course fee.
Module overview
Ten modules. Every module ends with a lab; open any module in the full curriculum below for its topics.
| # | Module | What you cover |
|---|---|---|
| 1 | Generative AI and Claude Fundamentals | How LLMs work, which Claude model to pick, and your first Claude API call. |
| 2 | Prompt Engineering and Context Engineering | Write prompts that hold up in production, and cut cost with prompt caching. |
| 3 | Structured Outputs, Tool Use and Function Calling | Give Claude tools and dependable JSON so it can act, not just answer. |
| 4 | Retrieval-Augmented Generation (RAG) and Knowledge Grounding | Ground answers in your own documents with embeddings, retrieval and citations. |
| 5 | Model Context Protocol (MCP) | Connect Claude to internal systems with Model Context Protocol servers. |
| 6 | Agent Architecture and Design Patterns | Choose workflows versus agents, and implement the patterns that scale. |
| 7 | Building Agents with the Claude Agent SDK | Build agents that read files, run commands and search the web, safely. |
| 8 | Multi-Agent Systems, Skills and Claude Code | Split work across specialist subagents, and use Claude Code on real repositories. |
| 9 | Evaluation, Safety, Observability and Deployment | Measure quality, harden against prompt injection, and ship to production. |
| 10 | Capstone Project and Wrap-up | Ship one complete agent, then present it and get reviewed. |
About the Program
This is a Claude API course first: you start with how the models behave and how to call the Messages API, then build the three capabilities production AI applications need — retrieval over your own data, tools the model can call, and agents that plan and finish multi-step work.
It is also a practical LLM application development course. Roughly 40% is concepts and demos, 60% is hands-on labs in Python or TypeScript, and the final module is a capstone you architect, build and present.
What You Will Build
Four working artifacts, each built in a live lab rather than demonstrated. By the end you have a portfolio you can show in an interview or reuse at work.
A RAG assistant
A document Q&A assistant over a company knowledge base that retrieves the right passages and answers with citations, so every claim traces back to a source.
An MCP server
A custom Model Context Protocol server that exposes an internal API as tools and resources, then connected to Claude and to the Agent SDK.
An AI agent
A research agent that plans, calls tools, reads sources and writes a cited report, plus a coordinator that delegates to specialist subagents.
A capstone project
One complete agent for support, analytics, engineering or operations: architected, built, evaluated and demoed in the final module.
Your trainer
Trainer: Kiran Kumar
Trainer, Generative AI with Claude
- Teaches the full program, from Claude fundamentals through to agentic systems and deployment.
- Recorded the orientation webinar on this page, walking through the course flow and what you will build.
Orientation with the Trainer
Learn more about the generative AI course directly from the trainer himself. He introduced the core concepts during this weekly webinar.

Weekly Webinar Introduction
In this session, the trainer walks through the course flow, the core concepts of Generative AI with Claude, and what you'll be able to build by the end of the program.
Watch the full webinarFull curriculum
Ten modules of three hours each. Expand a module to see its topics and lab.
- How LLMs work: tokens, context windows, sampling, temperature
- The Claude model family and how to pick a model for cost, speed and capability
- Anthropic's approach to safety and reliability, and what it means for builders
- Ways to use Claude: the Claude app, Claude Code, API and Console
- Lab: Set up the Console, create API keys, make your first Messages API call in Python and TypeScript
- Lab: Experiment with system prompts, multi-turn conversations and streaming
What You'll Learn
- Explain how LLMs work and choose the right Claude model for cost, speed and capability
- Design effective prompts with structured outputs, reasoning techniques and prompt caching
- Build the tool-use loop that turns an LLM into an agent, with validated JSON outputs
- Create grounded answers with RAG, citations and agentic retrieval
- Build and connect custom MCP servers that expose internal APIs to Claude
- Apply agent design patterns and build agents with the Claude Agent SDK
- Evaluate, secure, observe and deploy agents in production
- Ship a complete capstone agent for support, analytics, engineering or operations
Who Should Attend
- Developers and engineers with basic Python or TypeScript
- QA engineers and testers moving into AI-powered automation
- Technical professionals familiar with REST APIs and JSON
- Builders ready to move from prompting to full agentic AI systems
This is not a data science or model-training program: there is no maths-heavy theory and no model fine-tuning work. It is a practitioner course on building applications with an LLM.
Course Format & Schedule
Live, hands-on sessions
Ten live modules, with a lab in every module. About 40% concepts and demos, 60% hands-on labs, and all sessions run live online with the trainer.
Pick any free demo session below to meet the trainer and see the teaching style. All times are shown in PST and IST.
Free Sessions
| Date (PST) | Time (PST) | Date (IST) | Time (IST) |
|---|---|---|---|
| Mon, Nov 16, 2026 | 6:00 – 7:00 PM | Tue, Nov 17 | 7:30 – 8:30 AM |
| Tue, Nov 17, 2026 | 6:00 – 7:00 AM | Tue, Nov 17 | 7:30 – 8:30 PM |
| Tue, Nov 17, 2026 | 6:00 – 7:00 PM | Wed, Nov 18 | 7:30 – 8:30 AM |
| Wed, Nov 18, 2026 | 6:00 – 7:00 AM | Wed, Nov 18 | 7:30 – 8:30 PM |
| Thu, Nov 19, 2026 | 6:00 – 7:00 PM | Fri, Nov 20 | 7:30 – 8:30 AM |
| Fri, Nov 20, 2026 | 6:00 – 7:00 AM | Fri, Nov 20 | 7:30 – 8:30 PM |
Paid Training Dates
| Date (PST) | Time (PST) | Date (IST) | Time (IST) |
|---|---|---|---|
| Mon, Nov 30, 2026 | 6:00 – 7:00 PM | Tue, Dec 1 | 7:30 – 8:30 AM |
| Tue, Dec 1, 2026 | 6:00 – 7:00 PM | Wed, Dec 2 | 7:30 – 8:30 AM |
| Wed, Dec 2, 2026 | 6:00 – 7:00 PM | Thu, Dec 3 | 7:30 – 8:30 AM |
| Thu, Dec 3, 2026 | 6:00 – 7:00 PM | Fri, Dec 4 | 7:30 – 8:30 AM |
| Fri, Dec 4, 2026 | 6:00 – 7:00 PM | Sat, Dec 5 | 7:30 – 8:30 AM |
| Mon, Dec 7, 2026 | 6:00 – 7:00 PM | Tue, Dec 8 | 7:30 – 8:30 AM |
| Tue, Dec 8, 2026 | 6:00 – 7:00 PM | Wed, Dec 9 | 7:30 – 8:30 AM |
| Wed, Dec 9, 2026 | 6:00 – 7:00 PM | Thu, Dec 10 | 7:30 – 8:30 AM |
| Thu, Dec 10, 2026 | 6:00 – 7:00 PM | Fri, Dec 11 | 7:30 – 8:30 AM |
| Fri, Dec 11, 2026 | 6:00 – 7:00 PM | Sat, Dec 12 | 7:30 – 8:30 AM |
| Mon, Dec 14, 2026 | 6:00 – 7:00 PM | Tue, Dec 15 | 7:30 – 8:30 AM |
| Tue, Dec 15, 2026 | 6:00 – 7:00 PM | Wed, Dec 16 | 7:30 – 8:30 AM |
| Wed, Dec 16, 2026 | 6:00 – 7:00 PM | Thu, Dec 17 | 7:30 – 8:30 AM |
| Thu, Dec 17, 2026 | 6:00 – 7:00 PM | Fri, Dec 18 | 7:30 – 8:30 AM |
| Fri, Dec 18, 2026 | 6:00 – 7:00 PM | Sat, Dec 19 | 7:30 – 8:30 AM |
Claude vs Other LLMs
Claude is the model used in the labs because the course covers Claude-specific features: the tool-use loop, prompt caching, the Claude Agent SDK and MCP. The architecture and patterns you learn transfer to other providers, so teams on GPT or Gemini can apply the same skills.
| Model family | Where it is used | Strengths worth knowing | Considerations |
|---|---|---|---|
| Claude (Anthropic) | Claude apps, Claude Code, Claude API, Agent SDK | Strong long-form reasoning and instruction following, dependable tool use, long context, safety-focused defaults. | Model choice, limits and pricing differ by model; check current Anthropic documentation for your workload. |
| GPT models (OpenAI) | ChatGPT, OpenAI API, Assistants tooling | Broad ecosystem and third-party tooling, mature function calling, wide availability of tutorials and libraries. | Capabilities and pricing vary by model tier, so pin the model you evaluate against. |
| Gemini (Google) | Gemini app, Vertex AI, Google Cloud and Workspace | Native multimodal input and tight integration with Google Cloud data services. | Most convenient when your data and deployment already live on Google Cloud. |
| Open-weight models (Llama, Mistral, Qwen) | Self-hosted or through cloud inference providers | Deployment control, fine-tuning and predictable hosting economics for high-volume workloads. | You own the infrastructure, evaluation, safety and upgrade work. |
This is a starting point, not a ranking: the right model depends on your data, latency budget and evaluation results. iTeLearn training is independent and not affiliated with Anthropic, OpenAI or Google, and product names are trademarks of their respective owners.
Pricing & Enrollment
$199 early bird with code GENAI-199 · $299 regular, in USD.
Generative AI with Claude
Anthropic API credits are not included: the labs run on your own Anthropic account, so usage is billed to you by Anthropic.
iTeLearn training is independent and not affiliated with Anthropic.
Register Now
Save your seat. The paid program runs 15 one-hour live sessions, starting 30 November 2026.
Register for the SessionJoin our WhatsApp community for course enquiries and updates.
Generative AI with Claude: frequently asked questions
What is the Claude API?
The Claude API is Anthropic's programmatic interface to Claude models. This Claude API course calls it from Python and TypeScript to build applications: prompts and workflows, tool-using assistants, retrieval over your own documents, and agents that complete multi-step tasks.
What is MCP and why is it part of the course?
Model Context Protocol (MCP) is an open protocol, originally introduced by Anthropic, that standardises how AI applications connect to external tools and data. Instead of a custom integration per system, you run an MCP server that exposes tools, resources and prompts. The MCP course module has you build a server and connect it to Claude.
What are the prerequisites?
Basic Python or TypeScript, REST and JSON knowledge, and an Anthropic Console account with API credits. No previous AI or machine-learning experience is required, and the course is not a data science or model-training program.
How is the course scheduled?
The course is delivered as ten live modules, each with a hands-on lab. Free demo sessions run on 16, 17, 18, 19 and 20 November, and the paid program runs from 30 November to 18 December 2026, with 15 one-hour sessions starting at 6:00 PM PST. All sessions are live online with the trainer.
Do I get a certificate of completion?
Yes. You receive a certificate of completion after finishing the module labs and presenting your capstone project. It is a certificate of completion from iTeLearn, not an Anthropic certification.
How much does it cost?
The regular price is $299. The early-bird price is $199 with code GENAI-199 at checkout, in USD for learners in the USA, India and elsewhere.
Are Anthropic API credits included?
No. You use your own Anthropic Console account with API credits, so usage is billed to you by Anthropic. The course fee covers the live training, labs, assignments with feedback and capstone review.
Does the course cover RAG?
Yes. A full module of this RAG course covers embeddings, chunking, vector stores, hybrid search, reranking, citations and agentic retrieval, followed by a lab where you build a document Q&A assistant with citations.
Is iTeLearn affiliated with Anthropic?
No. iTeLearn training is independent and not affiliated with Anthropic. Claude and other product names are trademarks of their respective owners.
Refunds and cancellations
Cancel up to 7 days before the program start date for a full refund. Within 7 days of the start date you can move your fee to another iTeLearn program as a credit. If iTeLearn cancels or reschedules a program, choose a full refund or a credit. Missed live sessions do not qualify for a refund, because recordings and materials are shared with registered participants.
Read the full refund and cancellation policy →Other live programs
Every iTeLearn program is taught live by a practitioner, with recordings and materials for registered participants.
Live program
HireCraft - Master the Skills to get Hired
Live online job search coaching for software engineers in the USA and Canada.
View courseLive program
AI in Industrial Automation
Five sessions on a real production line, from process map to pilot.
View courseLive program
Tricentis Tosca Automation
Maintainable, model-based Tosca automation on a real-world app.
View course