A quick glance through this year’s NorDevCon schedule would give you the following brief, but insightful description of Maryleen’s talk:

The rapid evolution of AI, especially with the emergence of Large Language Models (LLMs), has been a transformative force in the tech landscape. Developers are enthusiastic about the potential of LLMs, but integrating them into real-world applications can be a daunting task as they often grapple with complex workflows, data challenges, and the lack of readily available tools. In this talk, I will discuss how a framework like LangChain simplifies LLM integration and redefines how we harness generative AI for tasks like report generation, data analysis, and even personalized customer service.

That being said, we decided to go a step further and spoil our community (as well as, hopefully, future audience!) with a bit more of a sneak-peak into the upcoming talk by interviewing Ms Amaizu, and encouraging her to give us a bit more of an insight on her upcoming speech!

---

🤖 What inspired you to choose this particular topic for your talk?

Maryleen: The rapid evolution of AI, especially with the emergence of Large Language Models (LLMs), has been a transformative force in the tech landscape. Developers are enthusiastic about the potential of LLMs, but integrating them into real-world applications can be a daunting task as they often grapple with complex workflows, data challenges, and the lack of readily available tools. In this talk, I will discuss how a framework like LangChain simplifies LLM integration and redefines how we harness generative AI for tasks like report generation, data analysis, and even personalized customer service.

🤖 How would you say your talk would benefit the developers, or professionals in other fields?

Maryleen: Software developers and programmers, especially those dealing with AI and language models, will gain valuable insights into overcoming the complexities associated with LLM-powered applications. LangChain's unified API and toolkit address challenges like model integration, context management, and the use of multiple LLMs. Data engineers, AI enthusiasts, and professionals seeking to leverage generative AI for diverse applications such as natural language processing, chatbot development, and data retrieval will find the talk rewarding.

🤖 If you could give us 1-3 main takeaways for your upcoming talk, what would they be?

Maryleen:

Use LLM framework to automates workflows and eliminates manual data manipulation, significantly reducing operational costs and human effort.

The dynamic nature of "chains" and "agents" allows you to refine outputs, and continuously learn, keeping your processes relevant and efficient.

Short-term and long-term memory addition to LLMs can improve context handling.

🤖 Are there any specific tools, technologies, or frameworks that you will be discussing in your talk?

Maryleen: Yes, the talk will focus on LangChain as the central framework. It will delve into its architecture, components such as Model I/O, Data Connection, Chains, Memory, Callbacks, and Agents. Supported LLMs from OpenAI, as well as the integration with vector databases and data retrieval modules.

🤖 What kind of audience engagement do you prefer, if any?

Maryleen: I'm open to questions midway through and after the talk.

🤖 Are there any “teasers” you could give us?

Maryleen: I'm particularly enthusiastic about discovering how to streamline the complex process of working with LLMs and delving into the future possibilities of frameworks that can support these endeavors.

---

We, of course, encourage you to meet Maryleen up close and personal at the upcoming NorDevCon 2024, but if fate has it otherwise, you can also check out her work and get in touch with her via LinkedIn.

If you haven't already, you can, of course, get the tickets to see Maryleen and other speakers at our Conference.

Thanks, Mez, Alex, Ida, & The nor(DEV):con Team