Build autonomous AI agents that plan, use tools, and orchestrate multi-step workflows — from single-agent loops to production multi-agent systems with observability.
The Agentic AI course in Nagpur teaches AI agent architectures, tool use and function calling, planning and memory systems, multi-agent orchestration with CrewAI, AutoGen, and LangGraph, the Model Context Protocol (MCP), observability and evaluation, and building autonomous production workflows. It is designed for developers and AI engineers who have LLM API experience and want to build agents that execute complex, multi-step tasks without human intervention.
Solid Python programming and working knowledge of LLM APIs (OpenAI or Gemini). Completing the Generative AI course or equivalent self-study is strongly recommended.
LangChain AgentExecutor: wrapping tools and the underlying loop
PythonOpenAI APILangChain Agents
Outcome: Build a ReAct agent loop from scratch that reasons about a task and selects tools iteratively until it reaches a correct answer — without a framework hiding the internals.
Module 2Tool Use & Function Calling
Week 2
OpenAI function calling: schema definition, tool selection, result handling
Tool design principles: single responsibility, typed inputs/outputs, idempotency
Building custom tools: web search (Tavily), Python REPL, file read/write, HTTP calls
Parallel function calling: handling multiple simultaneous tool requests
Error handling in tool results: graceful degradation strategies
Security considerations: what tools should and should not be able to do
OpenAI Function CallingPydanticTavily Search APIPython subprocess
Outcome: Build a custom toolbox of 4+ tools (search, calculator, file I/O, HTTP) with Pydantic-typed schemas, and wire them into an agent that selects and calls tools autonomously.
Plan-and-Execute agent: the planner model vs the executor agent split
Short-term memory: in-context conversation history management
Long-term memory: semantic memory with vector stores, episodic storage
External memory with Zep and mem0: storing and retrieving user context
Memory compression: summarising long conversation histories automatically
Handling context window limits in long-running agentic tasks
LangChain MemoryRedisZepmem0
Outcome: Build an agent with both short-term conversation memory and long-term semantic memory — one that remembers facts across sessions and plans multi-step tasks explicitly.
Module 4CrewAI — Role-Based Multi-Agent Systems
Week 4
CrewAI concepts: Crew, Agent, Task, Process (sequential vs hierarchical)
Defining agent roles, backstories, and goals for specialised behaviour
Task delegation: sequential task pipelines and task context passing
Hierarchical process: a manager agent assigns tasks to specialist agents
Inter-agent communication: how agents share output as context
Custom CrewAI tools and integrating third-party APIs as tools
Building a research and writing crew: researcher, analyst, writer agents
CrewAIPythonSerper APIDALL-E (optional)
Outcome: Orchestrate a 3-agent CrewAI crew (researcher, analyst, writer) that autonomously researches a topic, synthesises findings, and produces a structured report.
Module 5LangGraph — Graph-Based Agent Workflows
Week 5
Why LangGraph: stateful, cyclical graphs vs sequential chains
LangGraph fundamentals: StateGraph, nodes, edges, and conditional routing
Defining agent state: TypedDict state schema for shared memory
Conditional edges: routing agent flow based on tool output or score
Human-in-the-loop: interrupt before tool calls for human approval
Persistence and checkpointing: resuming interrupted agent runs
Subgraphs and modular agent composition in LangGraph
Outcome: Design and implement a stateful LangGraph workflow with conditional branching, human-in-the-loop interrupts, and run persistence — demonstrating production agent architecture.
Module 6AutoGen & Multi-Agent Conversations
Week 6
AutoGen ConversableAgent: user proxy agents and AI assistant agents
GroupChat: orchestrating multiple agents in a round-robin conversation
GroupChatManager: controlling who speaks next in a multi-agent session
Code execution agents: having agents write and run Python code autonomously
Nested chats: agents triggering sub-conversations to complete sub-tasks
AutoGen vs CrewAI vs LangGraph: a comparative analysis for team selection
Building a code review pipeline with a coder agent and critic agent
AutoGen (Microsoft)PythonCode ExecutorGroupChat
Outcome: Build an AutoGen GroupChat with a coder agent, a critic agent, and a user proxy — where the agents collaboratively write, review, and refine Python code autonomously.
What is MCP: the open protocol for connecting AI models to external tools and data
MCP architecture: host, client, server — and how they communicate
Building an MCP server in Python: exposing resources, tools, and prompts
MCP tools vs function calling: similarities and key differences
Connecting an MCP server to Claude Desktop and other MCP-compatible clients
MCP resources: serving dynamic context (database records, files) to AI models
Security model in MCP: capability permissions and sandboxed execution
MCP Python SDKClaude API (MCP host)FastAPIVS Code MCP extension
Outcome: Build a working MCP server in Python that exposes 2+ tools and 1+ resource, and connect it to a Claude or compatible AI host for tool-augmented AI interactions.
Outcome: Deploy an observed, evaluated, production-ready autonomous agent workflow — with LangSmith or AgentOps tracing, an LLM-as-judge evaluation suite, and a live public endpoint.
Tools & Technologies You Will Master
PPython 3.12
OOpenAI API
LLangChain
LLangGraph
CCrewAI
AAutoGen
MMCP Python SDK
PPydantic
mmem0 / Zep
LLangSmith
AAgentOps
FFastAPI
RRedis
DDocker
RRailway / Render
Language Framework Database DevOps AI Platform
Capstone Projects — What You Will Ship
Autonomous Research & Report Agent
Project 1
Problem: Build an agent that accepts a research topic, autonomously searches the web, reads sources, synthesises findings across multiple documents, and produces a structured written report — all without human intervention.
You ship: A deployed API endpoint that accepts a topic string and returns a markdown research report — with full LangSmith trace showing every agent step, token count, and tool call.
Multi-Agent Software Development Crew
Project 2
Problem: Orchestrate a CrewAI or AutoGen crew where a Product Manager agent writes user stories, a Developer agent writes code to fulfil them, and a QA agent reviews and tests — simulating a mini software team.
CrewAI or AutoGenOpenAI GPT-4oPython Code ExecutorAgentOps
You ship: A runnable multi-agent pipeline that takes a feature request and outputs working Python code, test cases, and a QA review — with AgentOps session recording shared as portfolio evidence.
Intelligent Automation Agent with MCP
Project 3
Problem: Build an MCP server exposing tools for a business use case (e.g., reading a CRM database, sending emails, looking up inventory) and wire it to an AI client that executes business workflows autonomously.
You ship: A production MCP server with 3+ tools and 2+ resources, integrated with Claude or a LangGraph agent, deployed and documented — demonstrating real MCP architecture knowledge.
Career Outcomes — Where This Course Takes You
Target Role Titles
AI Agent Engineer
Agentic AI Developer
AI Automation Engineer
LLM Systems Engineer
AI Product Engineer
Senior AI Engineer (with GenAI experience)
Skills You Will Reach
Skills matrix for Agentic AI
Skill
Level
Indicator
Agent Architecture Design
Advanced
Tool Use & Function Calling
Advanced
LangGraph Stateful Workflows
Proficient
CrewAI Multi-Agent Orchestration
Proficient
Model Context Protocol (MCP)
Proficient
Agent Memory & Planning
Proficient
Observability & Evaluation
Proficient
AutoGen Code Execution Agents
Foundational
Sample Interview Questions You Will Be Ready For
1“What is the difference between a chain and an agent in LangChain?”
2“Explain the ReAct loop: how does an agent decide when to use a tool vs generate a final answer?”
3“How does LangGraph differ from CrewAI for orchestrating multi-agent workflows?”
4“What is the Model Context Protocol (MCP) and how does it differ from OpenAI function calling?”
5“How would you debug an agent that keeps looping and never reaches a final answer?”
Salary benchmarks: <<PLACEHOLDER: Average salary for AI Agent Engineers in India — source: AmbitionBox / Glassdoor India / LinkedIn Jobs India, verified [month year]>>
Why Skilleze Technologies — Not Just Another Institute
Skilleze Technologies vs generic institutes comparison
What You Get
Skilleze Technologies
Generic Institutes
Mentorship
1-on-1 mentor sessions + PR code reviews
Recorded videos only
Projects
2–3 deployed capstone projects with live URL
Toy exercises or no projects
Syllabus currency
Updated to 2025–26 industry tooling
Often 2–3 years behind market
Batch size
Small cohorts — mentor access is real
Large batches, no individual feedback
Certificate verifiability
Verifiable reference number issued
No verification mechanism
Location
Offline Nagpur + online + hybrid
Online only or fixed city only
Career layer
ATS resume + LinkedIn + mock interviews
Not included
Comparisons are factual statements about our own offering. We do not name or endorse any specific competitor.
Frequently Asked Questions — Agentic AI
What is Agentic AI and is it different from Generative AI?
Generative AI refers broadly to AI that generates content — text, images, code. Agentic AI is a subset where AI systems autonomously plan, use tools, and take multi-step actions to complete a goal without constant human guidance. Think of Generative AI as the brain — Agentic AI adds hands, memory, and the ability to work across sessions to accomplish complex tasks.
What is the fee for the Agentic AI course in Nagpur?
Fees are shared during a one-to-one advisor call. EMI plans and batch-specific pricing are available. Connect via WhatsApp or the enquiry form — an advisor will provide the full fee structure within one business day.
Do I need to complete the Generative AI course first?
Strongly recommended. The Agentic AI course assumes comfort with LLM APIs, embeddings, and RAG concepts. If you have equivalent self-study experience — you've built RAG pipelines and used OpenAI function calling — you can join directly. Speak to an advisor to assess your readiness.
What is CrewAI and why does this course use it?
CrewAI is an open-source framework for orchestrating multiple AI agents with defined roles, goals, and task assignments — similar to a team of specialists. The course teaches CrewAI because it produces readable, maintainable agent code and is actively used in production AI applications. LangGraph and AutoGen are also covered for comparison.
What is the Model Context Protocol (MCP)?
MCP (Model Context Protocol) is an open standard created by Anthropic that defines how AI models connect to external tools, databases, and resources. It allows AI agents to access any data source or tool through a standardised interface — similar to how HTTP standardised web communication. MCP knowledge is increasingly valuable as it gains industry adoption in 2025–2026.
Is Agentic AI a viable career path in Nagpur?
Agentic AI engineering is one of the highest-demand specialisations in tech globally. The supply of engineers who can build, evaluate, and deploy reliable agents is extremely low relative to demand. In Nagpur, the skills are transferable to remote work for global companies and AI-first startups — making it one of the most location-independent career paths available.
Is this Agentic AI course available offline in Nagpur?
Yes. Offline classes at our Wardha Road centre in Nagpur, with online live and hybrid options. Weekend batches run Saturday–Sunday for working professionals.
How is this course different from the Generative AI course?
The GenAI course focuses on RAG, embeddings, APIs, and single-LLM features. This course assumes those skills and teaches agent architecture, multi-agent orchestration, tool use, MCP, and observability — the engineering discipline of building systems where AI does complex work autonomously over multiple steps, not just a single prompt-response.
Will I get hands-on project experience in the Agentic AI course?
Yes. Every week includes a project deliverable — from a bare-metal ReAct agent to a deployed multi-agent crew to an MCP server. The capstone is a fully observed, evaluated, deployed autonomous workflow agent with LangSmith or AgentOps tracing — a significant portfolio piece.
