• Home
  • /
  • Agentic AI Engineering: Planning, Tool Use & Multi-Agent Orchestration
Find a workshopPay feeRequest a callbackVerify a certificate
AI · Moderate · Video-based learning

Agentic AI Engineering: Planning, Tool Use & Multi-Agent Orchestration

Build AI Agents That Plan Tasks Use Tools and Collaborate to Solve Real-World Problems

★ 4.1 (31 ratings)Access 24/7 · 365 daysVideo lectures of 3+ hrsCertificate
Registration open

About This Course

This three-day hands-on workshop introduces the design and implementation of agentic AI systems using Python, LangChain and LangGraph. Participants progressively build a tool-enabled agent, add context management and retrieval, and develop a small multi-agent workflow. Practical sessions cover structured outputs, error handling, human approval and performance evaluation through a guided research assistant project.

Aim

To equip participants with practical skills to build AI agents that plan tasks, use tools and collaborate through reliable Python-based workflows.

Course Objectives

  • Understand agent architectures, task planning and execution loops.
  • Integrate Python functions and document retrieval as agent tools.
  • Manage context and workflow state across multiple steps.
  • Coordinate specialist agents using a supervisor workflow.
  • Apply error handling, human review and basic evaluation.
Next: Syllabus →

Syllabus

📅 Day 1: Agent Foundations, Task Planning & Tool Use

Focus: Build an agent that selects and uses tools to complete a task.

  • Understand AI agents, predefined workflows and agent execution loops.
  • Break user requests into steps with clear completion criteria.
  • Define Python tools with structured inputs and outputs.
  • Connect an agent to a document search tool and calculator.
  • Apply tool-call limits and handle basic execution errors.

🛠️ Hands-on:
Build a single agent that searches a supplied document collection, performs calculations and produces a structured answer.

🧰 Tools: Python, Jupyter Notebook, LangChain, Pydantic and an LLM API.

📅 Day 2: Context, Retrieval & Reliable Workflows

Focus: Maintain task context and build workflows that can recover from failures.

  • Manage conversation history, task state and relevant context.
  • Introduce retrieval-augmented generation using supplied documents.
  • Build conditional workflows with LangGraph.
  • Implement checkpoints, retries and stopping conditions.
  • Add human approval before a selected tool action.

🛠️ Hands-on:
Extend the agent into a document research workflow that retrieves evidence, maintains state and pauses for user review.

🧰 Tools: LangGraph, local document retrieval, checkpointing and Python logging.

📅 Day 3: Multi-Agent Orchestration, Evaluation & Capstone

Focus: Coordinate specialist agents and assess the quality of their outputs.

  • Compare single-agent and multi-agent designs.
  • Assign researcher, analyst and reviewer responsibilities.
  • Implement a supervisor workflow with controlled delegation.
  • Share structured results between agents.
  • Evaluate task completion, evidence support, tool errors, latency and token usage.
  • Inspect execution traces and identify unnecessary agent steps.

🛠️ Hands-on Capstone:
Build a small AI research team that searches supplied documents, analyses a curated CSV dataset and prepares an evidence-supported report for human review.

🧰 Tools: LangGraph, Pandas, Pydantic and execution tracing.

Who Should Enrol?

  • Undergraduate and postgraduate students in Computer Science, AI, Data Science and related disciplines.
  • Research scholars, academicians and professionals interested in AI-assisted research and automation.
  • Software developers and data analysts seeking practical agent development skills.

Prerequisites: Basic Python programming, familiarity with generative AI and access to the selected LLM API. Prior experience building AI agents is not required.

Course Outcomes

  • Build an AI agent with structured tool inputs and outputs.
  • Create a document research workflow with source-supported responses.
  • Implement checkpoints, execution limits and approval steps.
  • Coordinate research and analysis agents to generate a report.
  • Assess task completion, tool errors, response quality and resource usage.
Next: Fees →

Course fee

Learners in India

₹2,499

Learners outside India

$50

  • Video lectures
  • Study material
  • Certificate after completion
  • 365-day access

Enrol in the course

The recordings are being uploaded. They appear in your dashboard as soon as they are ready, and we will e-mail you then.

Next: FAQ →

Frequently asked questions

How is this course taught?

Through recorded video lectures from the original live workshop, with the study material. Watch at your own pace, any time.

How long do I have access?

365 days from your registration, on your NanoSchool dashboard.

Do I get a certificate?

Yes, an e-certificate after completion, with a QR code anyone can verify and an "Add to LinkedIn" button.

How can I pay?

By card, UPI or PayPal on the payment page; international bank transfer is also available there.

What is the refund policy?

See our Returns, Refunds & Cancellation policy: Returns, Refunds & Cancellation policy

Need Help?

We’re here for you!


(+91) 120-4781-217
 WhatsApp +91 99581 61117

Enrol
Now
Explore

Registration open

More upcoming workshops

Live, mentor-led and certified. Dates and times in IST.

See all workshops
Visit Course Page
Support