Gaming Technology: Creating Lifelike Non-Player Characters and Procedural Tasks

This initiative involves the integration of "Roleplaying Robots" powered by Large Language Models (LLMs) to prompt-engineer unique personalities for fictional characters(NPCs). Additionally, we will try to dynamically create preset responses from the player to continue the conversation, fostering immersive and engaging conversations. The LLM will be able to give the player dynamic “Tasks” from the tests enhancing the gaming experience. We will use Natural Language Processing to mine the text for a procedural task-based robot that the players in any game can use to accomplish tasks.

We’ll have discussion groups, workshops, and walk-throughs to learn different tools and techniques to get ourselves familiarized with the project.

The current challenge in the gaming industry is the substantial investment, often exceeding $500 million, required for game development. This project aims to revolutionize low-budget game development by introducing procedurally generated tasks and artificial intelligence, ensuring a lifelike gaming environment. Beyond the conventional focus on character animation and world-building, we intend to leverage public domain "Lore" intellectual property (Not Copyright infringe) to create compelling narratives.

We’ll learn how to develop a machine-learning textual corpus model that trains on a specific character world that is public domain, and then query the robot with responses for tasks.

This same model could be applied to a copyrighted current lore system in a modern fictional work, but for the sake of practice, we will likely use something in traditional folklore.

• Develop a machine learning textual corpus model, adhering to copyright regulations, specifically trained on public domain fiction.
• Deploy LM Studio on the cloud with Cross-Origin Resource Sharing (CORS) capabilities to enable seamless API querying through a private LLM server using LM studio-Linux (beta) (Works in Ubuntu 22)
• Install KDE Desktop or Gnome to optimize the user interface experience.
• Explore techniques for custom knowledge integration into the LLM, with a specific emphasis on public domain fiction.
• Utilize the LM Studio AWS channel for efficient deployment.
• Establish a Text Mining Channel to extract pertinent information from the textual corpus.
• Implement a Bot Prompt Engineering Channel to design engaging and contextually relevant NPC responses.
• Develop procedural quest stage information through meticulous text mining.

1. How to work with cloud deployed prebuilt models 2. How to work with prompt engineering LLM's 3. How to work on Rest API's 4. How to mine text using NLP