Good day, everyone. Thank you for joining the session on Exploring AI Powered Chipboards and their Transformative Role in Platform Engineering. I'm Nagarajan Malone. I work for Intuit as a senior staff engineer, and I work on delivering AI driven interactions in my long career. I've looked far. In front teams and contributed to developer platforms, shipping tools, SDKs, and established self-serving kiosks. I always wondered if there is some intelligent assistant who can help me become more productive and efficient, wide, repetitive tasks. in this sta in this talk, I'm excited to talk about an intelligent assistant in the form of AI chat bots in platform engineering. AI powered chatbots are intelligent systems that can interact with users using natural language. They have the potential to significantly impact platform engineering, a critical area in modern software development that deals with infrastructure, automation, and DevOps practices. Today, we will dive into how these chatbots can revolutionize platform engineering by automating tasks, enhancing collaboration, and improving overall efficiency. AI powered chatbots are tools that use natural language, machine learning, and, generative AI to understand and respond to user commands in a conversational manner. These chatbots integrate with various backend systems, which again is enhanced by the machine learning or generative AI to perform complex tasks, making them powerful tools for platform engineering. An example is like chatbots. We already have chatbots, for, from Slack and from Microsoft Teams, which are, already being used for DevOps environments to streamline workflows and facilitate communications. And chatbots can be proactive. It can start the conversation based on a triggering event and is available 24 seven. what is the gaps? What are the gaps? the platform engineering, what are the gaps, that the chatbot can address? first, they simplify complexity. Managing complex infrastructural systems can be overwhelming, but chatbots provide an intuitive natural language interface that makes this task more accessible. Second, they automate repetitive tasks. Chatbots can handle routine operations like deployments, monitoring, and scaling, significantly reducing the manual workload for the engineers. And third, they enhance real time collaboration by integrating with communication platforms like Slack or Teams. AI chatbots enable teams to have actual real time conversations, which speeds up decision making and problem resolution. Last, they empower users. Chatterbots allow developers and engineers to perform tasks independently without needing specialized knowledge or waiting for central teams to step in. This self service capability speeds up workflows and enhances productivity. Transformative benefits of AI Chatterbots. Efficiency. By automating repetitive tasks, chatbots significantly speed up operations, leading to quicker deployments, faster issue resolution, and overall improved productivity. Scalability. AI chatbots can handle increasing workloads without performance bottlenecks, allowing platforms operations to scale efficiently as demand grows. Proactive operations. Chatbots can monitor systems in real time. automatically triggering responses to potential issues before they escalate, which helps in maintaining the health and stability of the platform. Consistency. Chatbots can ensure standardized execution of tasks, reducing variability and the risk of human error. This consistency is crucial in maintaining reliable and predictable platform performance. User empowered helps. One of the most impactful benefits is how chat bots empower engineers by providing them with the tools to manage tasks independently. This helps help reduces the bot link speeds up development cycles, and free out the security team to focus on more strategic initiatives rather than repeating, the same, work with different teams they support and, getting exhausted. Some of the, practical use cases, infrastructure provisioning, chatbots can automate the setup of new environments. For example, a developer can, instruct the chatbot to create a testing environment for two EC2 instances. And the chatbot will handle the entire process. the CNCD pipeline management chatbots. Can allow the developers to trigger paints, deployments, and even rollback, rollbacks via simple chat commands, making continuous integration and delivery processes more accessible and efficient. Monitoring alerts. Chatbots can integrate with monitoring tools to send real time alerts about system performance issues directly within the chat interface. Developers can respond immediately, running diagnostic commands or triggering recovery procedures right from the chat. Incident management. During outages or incidents, chatbots streamline response efforts by automating workflows. They can escalate issues, provide real time updates, and coordinate the actions needed to resolve quickly. Finally, knowledge access. AI chatbots can instantly retrieve documentation, facts, or troubleshooting steps, providing teams with the information they need to resolve issues quickly without having to search through the extensive manuals or databases, leveraging the power of GDT AI. An overview of the typical chatbot architecture. It has a user interaction layer. Chatbot interacts with the users via platforms like Slack, Teams, or other chat interfaces. It can also be in the build phase. Layers include a natural language processing engine that interprets the user commands. The core logic layer is essential. The, essentially, the brain of the chatbot. Here, the dialogue manage, dialogue management. task orchestration takes place, determining the appropriate action space around users input. The integration layer connects the chatbot to various tools and platforms such as, Jenkins, Kubernetes, or cloud management systems. this allows the chatbot to perform complex tasks like deployment or monitoring directly through these integrations. Security layer is essential, To ensure that all interactions are secure with, identity and access management controlling who can execute specific commands, ensuring only secure, only authorized users can perform critical operations. And finally, continuous learning layer is where the chatbot, AI models are maintained and improved over time. This layer allows the chatbot to learn from past interactions, improving its accuracy and functionality with each use. Here is a high level architecture diagram. There is a natural language, the chat, we see the chat interface, and then a natural language understanding component. And that component extracts the entities and then there is a dialogue management component. And the dialogue management component can be integrated with the backend systems. And then there is a, the response is passed to the general message generator and message generator sends the response to the chat thread. Here's a little more detailed architecture. You have a layer, we already saw that, the messaging layer, and then the NLB. The NLB receives internet recognition and entity recognition. Intent is the intent of the user, exactly extracting the actual intent. What is the work that needs to be done? What is the request coming from the user? Entity is what is that the work is going to, or is that, what is that going to get updated? Thank you. And then we have a Co AI layer. there's a dialogue management, which is actually retaining the previous conversations. And, the context and the memory. And, the knowledge based integration, the all the exhaustive stuff, documentations, the, which really I can scan through and then give a, give a response. And integration with a large language model. Okay. And we have the core layer, the response generation layer. The response generation layer can be template based. It can provide dynamic responses, or it can also provide a multimedia response, like it can generate a report and share it on the chat thread. Of course, we have a backend services, which stores the data. Analytics and monitoring can be built on top of it. And, then there's a security. And throughout all these layers, we want to ensure that there is compliance to all the, to all the regulations, regulatory bodies, and we ensure the privacy of the end user is also maintained. Challenges this, there are challenges that we need to be aware of. We need to be aware of while implementing a chat interaction. the managing the complex commands with the natural language understanding, it can be a task that needs to be done to Avoid any, ambiguous misinterpretation of the commands coming from the user. Security is another major concern since, chatbots can access uncontrolled sensitive systems, robust security measures like role based access control and multi factor authentication are essential to prevent unauthorized access. Integration can be challenging, especially in complex IT environments, which a chatbot needs to connect with various tools and systems. Ensuring compatibility and a seamless data flow between these integrations is crucial. User adoption is another key success factor for the chatbot. If it's not user friendly or trustworthy, it may be, or it may not be adopted widely. Intervening its effectiveness, ensuring the chatbot is intuitive and providing adequate training are essential steps to encourage. Adoption. Some of the emerging trends in the, trends in the AI driven chat bots is multimodality. So the input may not be always in the form of text. It can be a voice based command or it can be a document, or can be a gesture. So there are multiple ways base to encrypt data into the feed, the data to the h and bot. And you can also preserve. in the format expected by the user. And, deeply integrated integrations enables users to navigate out and into the chatbot seamlessly. For example, chatbot gets the context of the activities user has done outside of the chatbot, selection of a certain component in that selection of the converted interface, which is totally outside of the chatbot. the context about the user activity is passed to the chatbot and it readily gets the context and works seamlessly. In conclusion, AI powered chatbots are transforming platform engineering by automating tasks, enhancing efficiency, and improving overall user experience. This chatbots bridge critical gaps in complexity, automation, and real time responsiveness, making platform engineering more accessible and effective. As we move forward, it is worth considering how AI chatbots can be integrated into the platform engineering practices to drive innovation and operational excellence. Thank you. Thanks for your time. I would like to open the floor to any questions you might have.