10 Weeks Agentic AI Engineering Accelerated Training

The course progresses into advanced concepts of Agentic AI orchestration and control, focusing on how to design agents that can handle complex, multi-step tasks. Students explore the differences between frameworks like LangChain and LangGraph, understanding how each supports the development of scalable and structured AI systems. You will also learn how to model agent behavior using state machines, enabling better control over how agents transition between tasks and decisions.

This module introduces multi-step reasoning flows, allowing agents to break down complex problems into manageable steps, as well as routing and decision nodes that help agents choose the right path or tool based on context. These concepts are essential for building reliable, production-ready AI agents that can operate autonomously in dynamic environments.

Tech Stack: LangChain, LangGraph, Python, and orchestration frameworks for building structured AI workflows.

The course then explores memory and context management in Agentic AI, a critical component for building intelligent systems that can retain information and improve over time. Students learn the difference between short-term and long-term memory, and how agents use conversation memory to maintain context across interactions for more coherent and personalized responses.

You will also dive into vector embeddings and retrieval strategies, understanding how agents store, search, and retrieve relevant information efficiently. This module highlights how memory systems enhance an agent’s ability to reason, adapt, and provide accurate, context-aware outputs in real-world scenarios.

Tech Stack: Python, vector databases, embeddings APIs, and retrieval frameworks for building context-aware AI systems.

• The course then dives into Retrieval-Augmented Generation (RAG), a powerful approach for building AI agents that can access and use external knowledge in real time. Students learn the core RAG architecture and how it enhances model responses by combining language generation with relevant data retrieval, making outputs more accurate and context-aware.

  You will explore how to design a knowledge ingestion pipeline, including techniques for processing and preparing data, as well as effective chunking strategies to ensure information is stored and retrieved efficiently. The module also covers query optimisation, helping agents retrieve the most relevant data quickly and improve overall performance.

  This module equips you with the skills to build intelligent systems that can leverage large datasets, documents, and knowledge bases to deliver reliable, up-to-date insights.

  Tech Stack: Python, vector databases, embeddings models, and RAG frameworks.

The course then focuses on tool integration and external system interaction, enabling AI agents to move beyond text generation and take meaningful actions. Students learn how to design and manage a tool registry, allowing agents to dynamically select and use the right tools based on context and tasks.

You will explore how to build API-calling agents that can interact with external services, retrieve real-time data, and trigger automated workflows. The module also covers the use of web scraping tools and database integrations, giving agents the ability to gather information from the web and query structured data sources effectively.

This module is essential for building powerful, action-oriented AI agents that can operate in real-world environments and automate complex processes.

Tech Stack: Python, APIs, web scraping libraries, and database integration tools.

The course then introduces Agentic AI for AIOps, focusing on how intelligent agents can enhance IT operations through automation and real-time decision-making. Students gain an overview of AIOps and how AI agents can be applied to monitor systems, detect issues, and respond to incidents more efficiently.

You will learn how to design specialized agents such as incident triage agents that can prioritise and route issues, as well as log analysis agents that can process and interpret large volumes of system logs. The module also covers alert correlation, enabling agents to connect related events and reduce noise, leading to faster and more accurate problem resolution.

This module equips you with the skills to build AI-driven operational systems that improve reliability, reduce downtime, and support modern DevOps and SRE practices.

Tech Stack: Python, monitoring/logging tools, AI models for analysis, and automation frameworks.

The course then focuses on observability, evaluation, and reliability in Agentic AI, ensuring that intelligent systems are transparent, measurable, and trustworthy. Students learn how to implement agent tracing to monitor decision flows and understand how agents reason through tasks in real time.

You will explore prompt logging techniques to track inputs and outputs, enabling better debugging and continuous improvement. The module also covers evaluation metrics for assessing agent performance, along with methods for hallucination detection to identify and reduce inaccurate or misleading responses.

This module is essential for building robust, production-ready AI agents that can be monitored, evaluated, and continuously optimized for accuracy and reliability.

Tech Stack: Python, observability tools, logging frameworks, and evaluation libraries for AI systems.

The course then focuses on security, governance, and operational control in Agentic AI, ensuring that intelligent systems are safe, reliable, and cost-efficient. Students learn how to defend against prompt injection attacks, where malicious inputs attempt to manipulate agent behavior or extract sensitive information.

You will explore output validation techniques to ensure AI responses meet expected formats, accuracy standards, and safety requirements. The module also covers policy enforcement, enabling organizations to define and apply rules that govern how agents behave in different scenarios. In addition, students learn cost control strategies to manage API usage, optimize performance, and prevent unnecessary resource consumption.

This module equips you with the skills to build secure, compliant, and cost-effective AI agent systems suitable for real-world deployment.

Tech Stack: Python, AI safety frameworks, API governance tools, and monitoring systems.

The course then focuses on building and deploying Agentic AI systems on AWS, showing how to move from local prototypes to scalable cloud-native architectures. Students learn how to design end-to-end agent systems using AWS services that support computation, storage, retrieval, and orchestration.

You will explore how to structure agent architectures on AWS using API Gateway with Lambda or ECS for scalable execution. The module also covers integration with AWS Bedrock for foundation models, enabling powerful generative AI capabilities. In addition, students learn how to use OpenSearch as a vector store for semantic retrieval and S3 as a centralized knowledge base for storing documents and training data.

This module equips you with the skills to design production-grade AI agents that are scalable, secure, and cloud-ready.

Tech Stack: AWS Lambda, ECS, API Gateway, Bedrock, OpenSearch, S3, and Python for orchestration.

The course then focuses on Agentic AI MLOps and platform engineering, teaching students how to build, deploy, and manage AI agents at scale using modern DevOps practices. Students learn how to create CI/CD pipelines using Jenkins or GitHub Actions to automate the testing and deployment of AI agent systems.

You will explore Infrastructure as Code (IaC) using Terraform to provision and manage cloud resources consistently and efficiently. The module also covers model versioning, enabling teams to track, manage, and roll back different versions of AI models and agent configurations. In addition, students learn how to design multi-tenant agent platforms that can securely serve multiple users or organizations from a single scalable infrastructure.

This module equips you with the skills to operate Agentic AI systems like production-grade software products in real-world environments.

Tech Stack: Jenkins, GitHub Actions, Terraform, Python, AWS services, and model management tools.