Presentation on building a customer support bot with Langraff, Amazon Bedrock as base. LangChain Python library for LLMs. RAG - Retrieval Augmented Generation for accessing current information. LangChain using Bedrock and Amazon Knowledge Base. LangRaph classifies queries intelligently. Using LandGraph on top of LangChain for multiple scenarios with human intervention. Connected to Amazon Bedrock for knowledge base. No guessing with large language models; rely on a connected knowledge base for accurate responses. Chat GPT creation involves memory for context; tools like Land Graph aid in this process. Land Graph usage for flow creation; Demo with Kiro agent ID and new file uploads. Setting up AWS MCP server with agent skills; Using AWS agent toolkit for best practices in service setup; Flow for intent routing and order refund eligibility. Handling Order Refund Requests Using Agent Tools and Lookup; Case Sensitivity in Order Number Search; Validating Order Eligibility for Refund. Uploading Knowledge Base to AWS MCP Server; Verification of Email Address for Refund; Human in the Loop for Refund Approval. Human in the Loop Feature of LaneGraph; Uploading Documents to Bedrock; Utilizing Vectors for Knowledgebase Queries. Utilizing AWS Agent Toolkit for Workflow Best Practices; Connecting to KnowledgeBases for Querying; Building App with Mastering LaneGraph. Generating Design and Requirements for App Development; Utilizing Agent Skills in Spec-Driven Workflow; Ensuring Accuracy with Mastering Langraff Skill. Detailing Setup for Coding Assistant with AWS Agent Toolkit; Requesting Feedback and Interaction; Importance of Langraff in Graph-based Solutions. Enhancing Chatbot Prompt Suggestions Dynamically based on Conversation Context; Utilizing In-Memory Thread for Follow-up Questions in Langraff; Swapping Question Prompts with Tags for User-friendly Experience. Utilizing Different Models for Chatbot Tasks; Implementing Security Measures; Continuous Knowledge Base Updates for Accuracy.

Training engineers for proper AI usage without prompt AI dependency. Challenges in ensuring code quality with AI development. Importance of DRY principle in optimizing AI usage. Advocating for effective AI practices implementation with project-specific rules. Establishing AI rules and standards for seamless collaboration. Maximizing AI efficiency with agents mimicking human roles. Encouraging self-education, setting standards, and focusing on quality for AI proficiency. Emphasizing context, validation processes, and specialized agents for maximizing AI efficiency.

Anri Fisher, with over 12 years in tech, focuses on training tech leads and addressing time management challenges. Tech leads shift to a supportive role in empowering team members in growth discussions and one-on-one meetings. Overcoming feedback and delegation struggles through clear communication and expectations is crucial for effective leadership. Strategies for proper delegation include setting clear expectations, agreeing on timelines, and defining processes. Ensuring team alignment, shared timelines, and clear ownership are key for project success in tech teams.

Voice interaction is evolving rapidly with AI agents moving beyond chat windows. Achieving natural conversations with voice AI presents significant challenges, especially in meeting real-time interaction expectations. Advancements in real-time conversation technology and infrastructure are improving latency gaps in voice apps. Optimizing latency at each stage of voice agent processing is crucial for efficient development. Semantic turn detection and efficient function calling are key to optimizing user experience in Voice AI. Workflow management, cost management, signal recognition, and observability are vital aspects in voice AI development.

Daniel Avila discusses Cloud Code templates and components for AI coding, emphasizing the importance of proper configuration and incorporating skills. Control permissions, settings, commands, sub-agents, and skills are essential for efficient workflow in Cloud Code. Cloud Code development benefits from skills injecting context for AI agents and providing clarity in configurations. Managing skill invocation, model selection, and advanced skill configuration are crucial for optimizing Cloud Code capabilities. Efficiently utilizing file references, scripts, and skill context enhance the functionality of well-crafted skills. Community interaction on GitHub and effective skill sharing practices contribute to successful team collaboration in Cloud Code development.

Developers and QA engineer Wendy discuss the challenges in development and testing, using John's experience as an example. John encounters challenges with expired coupons affecting the checkout process and sprint deadlines. Developers and testers have complementary perspectives in software development; introducing the concept of QAI for enhanced quality assurance and collaboration. The QAI introduces a new perspective to code development, focusing on thorough alignment and error scanning for enhanced quality assurance. The importance of code alignment in development, the consequences of silent failures, and the necessity of handling potential hidden failures with a security-focused approach. AI builds fast and trusts everything, teaches suspicion, identifies trust gaps, and addresses accessibility failures for a secure code environment. Every team has patterns to anticipate, leveraging skills for immediate help, and focusing on effectiveness and protection in AI development. John and I collaborate using QAI to enhance code effectiveness, prioritize code alignment, fix contradictions, and address hidden failures effectively. Ship clean code timely, find bugs early to save time and trust. AI and QAI enhance speed and safety in development. Wendy emphasizes the importance of detecting bugs early and the impact on user trust and deadlines.

The Talk delves into the exploration of Agents on the Canvas, integrating LLMs with visual elements and spatial relations. It covers the development of the Agent Starter Kit for navigating Canvas tasks and introduces fairies as collaborative agents. The discussion includes managing tasks, utilizing Cloud Code for project collaboration, creating interactive UIs, and customizing UI components. The evolution of agentic canvas code, agent interactions, user actions, token usage, and data handling with LLMs are highlighted, emphasizing experimentation and optimization for SDK integration.