Presentation Session 5 Transition roadmap.pdf

Presentation Session 5 Transition roadmap.pdf

• 1.
  Transitioning to AgenticAutomation 6 Part webinar series by the UiPath Community March – April 2025
• 2.
  2 Meet the team! SatishPrasad Solution Consultant @ IRIS UiPath MVP Rohit Radhakrishnan Community Lead - Asia Pacific & Japan @ UiPath Mannoj Batra Vice President @ NatWest Group UiPath MVP Mukesh Kala RPA Manager @ Boundaryless Group UiPath MVP Vibhor Shrivastava Director Community Advocacy @ UiPath Andreea Tomescu Community Manager @ UiPath
• 3.
  3 6 Part Series Registerfor each session here
• 4.
  4 Summary of AI-PoweredAutomation Webinar Series 1. Introduction to Generative AI & UiPath Autopilot - Differentiated RPA, AI, GenAI, and Agentic AI. - Explored LLM integration with UiPath. - Hands-on with Autopilot for Everyone: Document summarization and claims automation. - Emphasis on PII handling, productivity packages, and real-world healthcare use cases. 2. Context Grounding & Vector Databases in UiPath - Unpacked the power of context-aware AI in automation. - Explained the role of vector databases and memory in improving LLM responses. - Hands-on: Created a chatbot with vector indexing using UiPath. - Covered security, role-based access, and grounding principles for accurate AI comprehension. 3. AI Trust Layer, Governance – Explainability, Security & Compliance - Covered SHAP & LIME for AI interpretability. - Deep dive into UiPath’s AI Trust Layer. - Discussed PII/PHI masking, content moderation, authentication, and regulatory adherence. - Explained ethical AI and governance best practices for responsible automation.
• 5.
  5 Summary of AI-PoweredAutomation Webinar Series 4. UiPath Agent Builder & Memory Tools - Introduced UiPath’s vision for agentic automation. - Learned how to build task-specific agents using the Agent Builder. - Implemented short- and long-term memory with vector DBs. - Hands-on: Created invoice-processing and onboarding agents with persistent memory. 5. Multi-Agentic Orchestration with UiPath Maestro - Showcased orchestration of multiple agents using UiPath Maestro. - Modeled agent collaboration using BPMN and pub/sub APIs. - Ensured secure agent communication. - Simulated complex workflows: Supply chain automation & e-commerce order fulfillment.
• 6.
  6 AI Agent LearningRoadmap for UiPath Developers Here’s a quick roadmap to get started: Level 1: Master UiPath & GenAI Learn advanced UiPath concepts (workflows, error handling, APIs). Explore UiPath GenAI and integrate tools like OpenAI into your workflows. Get hands-on with Prompt Engineering and data preprocessing. Level 2: Build AI Agents Understand AI agents and how they differ from RPA. Learn frameworks like LangChain and build simple agents with UiPath. Integrate memory and multi-agent collaboration into your workflows. Learn UiPath Agent Builder Or Any Tool of Your Choice. Level 3: Scale AI Agent Development Dive into advanced frameworks and reinforcement learning. Learn about AI ethics, governance, and real-world use cases. Why Make the Shift? AI agents help automate complex, cognitive tasks that traditional RPA can’t.
• 7.
  7 1. Master AdvancedUiPath Concepts Before diving into AI, ensure you have a solid grasp on UiPath’s advanced features. This includes: • Advanced workflows, error handling, and optimization. • Integration with APIs and external systems. • UiPath Document Understanding and AI Center. 2. Explore UiPath GenAI Capabilities Take advantage of UiPath's Generative AI (GenAI) features. Learn how to integrate OpenAI, GPT, and other LLMs into your automation workflows. Key skills include: • Using UiPath’s AI Center and pre-built AI models. • Building GenAI-powered workflows with a focus on efficiency and accuracy. 3. Learn Prompt Engineering for UiPath Prompt engineering is a critical skill when working with AI models. You'll learn: • How to craft prompts for GenAI activities. • Techniques like zero-shot, few-shot, and chain-of-thought prompting to enhance automation. • Fine-tuning outputs using parameters like temperature and token limits. 4. Data Handling and Preprocessing in UiPath AI agents thrive on structured, clean data. Learn how to: • Clean and preprocess data for AI workflows. • Use UiPath’s data scraping and extraction tools for efficient data gathering. 1. 5. API Integration and Automation Master API integration to connect with cloud-based AI services (e.g., OpenAI, Azure AI). This includes automating tasks using REST, GraphQL APIs, and building custom connectors. Level 1: Deepen Your UiPath Expertise and Explore UiPath GenAI
• 8.
  8 1. Introduction toAI Agents Get familiar with AI agents and how they differ from traditional automation. Key concepts include: • Understanding agent-environment interaction and decision-making. 2. Learn Agentic Frameworks Explore frameworks like LangChain, AutoGen, and CrewAI to build low-code AI agents. Learn how to compare these with UiPath’s automation capabilities. 3. Build Your First AI Agent Start with a simple AI agent by integrating external LLMs into your UiPath workflows. Learn how to combine RPA and AI for end-to-end solutions. 4. Basics of Agentic Workflow Design workflows tailored for AI agents: • Breaking tasks into logical steps. • Orchestrating multiple agents for complex workflows. • Implementing error recovery mechanisms. 5. Learn About Agentic Memory Explore short-term vs long- term memory in AI agents. Learn how to: • Use vector databases like ChromaDB, Pinecone for memory storage. • Integrate memory systems into UiPath for more adaptive workflows. 6. Multi-Agent Collaboration Learn strategies to build multi- agent systems that collaborate seamlessly. This includes agent communication protocols and dependencies. 7. Implement RAG (Retrieval-Augmented Generation) RAG is a powerful technique for enhancing AI agents’ ability to handle complex document understanding. Learn how to build agentic pipelines for context and retrieval tasks. Level 2: Transition to AI Agent Development
• 9.
  9 1. Advanced AgenticFrameworks Take your AI agent development skills to the next level by building custom agents using Python and exploring advanced frameworks. Understand how to integrate reinforcement learning and cutting- edge AI models (e.g., GPT-4, Claude). 2. AI Agent Deployment and Scalability Learn how to deploy AI agents into production environments: • Scale your workflows using cloud infrastructure. • Monitor and optimize agent performance for long-term success. 3. Ethics and Governance in AI Agents AI is powerful, but it needs to be governed responsibly. Learn to: • Address bias and fairness in AI agents. • Ensure transparency and accountability in agent decisions. • Stay compliant with evolving AI regulations. 4. Real-World Use Cases Learn from real-world applications of AI agents, such as: • AI agents for customer support, finance, healthcare. • Combining UiPath automation with AI agents for seamless end-to-end solutions. Level 3: Advanced AI Agent Development
• 10.
  10 Thank You Register foreach session here