Agentic AI That Pays: Driving ROI Through Cost Control & Governance

  • Aug 08, 2026
  • 09:30AM – 05:30PM

About the Workshop

NVIDIA CEO Jensen Huang recently emphasized that AI is entering the era of "Useful AI"—moving beyond content generation to autonomous systems that can execute tasks, make decisions, interact with enterprise systems, and generate measurable business outcomes. The challenge for most organizations is not adopting AI, but achieving ROI without creating runaway cloud, model, and implementation costs.

This workshop focuses on how to identify high-ROI agentic AI opportunities, build practical AI agents, and deploy them with strong governance and cost controls.

This workshop is tailored toward mid- to high-tier executives and managers from both technology and business teams. The goal is to understand ROI-led planning and implementation of Agentic AI systems.

Prerequisites

  • A Laptop with Powerpoint or Google Sheets access

Workshop Modules

  • Understand what Agentic AI is at its core. Code through a few code snippets, deployment models, and cost implications.
  • Anatomy of an Agentic AI System: planning, reasoning, acting, memory, tool usage, workflow orchestration, single vs multi-agent architectures
  • Take a deep dive into popular use cases of Agentic AI such as Software Engineering, Autonomous Cybersecurity Operations, Enterprise Search and Knowledge Automation, Customer Service & Back-office Operations
  • Understand how Agents can be integrated into your existing usage workflows and software products
  • Explore different modalities of agents: text, text + images, text + videos, images, videos, multi-modal
  • Explore how Agentic AI can reshape your business landscape, identify customer potential , and create value drivers

  • Live example of an organisation, where agents can help, what is the ROI, what is the cost, how much time will it take to implement and generate value
  • Formulate a clear AI strategy that aligns with your organisation goals, consider scalability, sustainability, and change management
  • Discuss best practices for implementing Agentic AI projects, including evaluation datasets, debugging, observability, continuous measurement and improvement.
  • Learn about the critical roles needed in an Agentic AI team and the key competencies required and explore strategies for upskilling existing teams and fostering a culture of continuous learning.

  • Common cost traps: Using GPT-5 for everything, Excessive context windows, Agent loops with no limits, Poor retrieval architecture, Uncontrolled API consumption
  • Cost Optimization Strategies: Small models first, Hybrid model architecture, Intelligent routing, Retrieval-Augmented Generation (RAG), Caching and memory optimization, Open-source vs proprietary models, When not to use AI
  • Agentic AI Architecture for the Enterprise: Enterprise reference architecture. Data governance. Security and compliance. Observability and monitoring, AI evaluation frameworks, and an AI governance checklist
  • What to do: Agent workflow, Cost estimate, ROI estimate, Governance plan
  • What to Avoid: AI science projects, Unbounded agent autonomy, Vendor lock-in, AI without measurable business outcomes
  • Participants leave with: Agentic AI opportunity assessment framework, Cost optimization checklist. AI governance template, 90-day implementation roadmap, Executive ROI calculator

Instructor

We use cookies essential for this site to function well. Please click to help us improve its usefulness with additional cookies. Learn about our use of cookies in our Privacy Policy & Cookies Policy.

Show details