Location
Roy O. West Library 119 (Roy Instruction Classroom)
Event Website
http://depauw.campuslabs.com/engage/event/12454497
Start Date
7-5-2026 2:15 PM
End Date
7-5-2026 2:30 PM
Presentation Type
Thesis
Description
Our brain has limited fuel: every time we make a decision, it costs fuel. A Personal Artificial Assistant will help preserve that fuel by handling secretarial tasks and scheduling a daily timetable for its user. Personal Assistant Agent (PAA) is an AI assistant running as a text-based Windows desktop application to help users in scheduling and give reminders like a secretary. Users interact with the PAA conversationally. This project is a system of reasoning and a modular memory pipeline that makes any Large Language Model (LLM) behave like a personal assistant. The Language model is isolated inside a function to keep the system’s logic from interfering with the model’s weight. When the user enters their question or command into a basic graphical user interface (GUI) made with tkinter, the system fetches the user input and creates a prompt so the LLM can summarize a general idea of what the user wants. Then, it fetches relevant data from the vector memory system and/or searches the internet for relevant information. The system sends another query to the LLM with the supporting context and instructions to get a response to the command using a Python script to update or manage the requested application (Google Calendar, alarm, or notepad). In short, the central system maintains the context window, memory management, and a dynamic backend prompt to get the model responses and uses those responses. The system operates entirely on local hardware and demonstrates how a chatbot can be made from public language models. Potential users include researchers, businessmen, salesmen, and students interested in having a secretary that tailors the schedule for you and does trivial tasks like scheduling meetings, alarms, or taking notes. The end goal of the project is a prototype that can do the tasks mentioned above through an application on a laptop.
See coding folder here: Visit Personal-Assistant on GitHub
Included in
Private Assistant Agent
Roy O. West Library 119 (Roy Instruction Classroom)
Our brain has limited fuel: every time we make a decision, it costs fuel. A Personal Artificial Assistant will help preserve that fuel by handling secretarial tasks and scheduling a daily timetable for its user. Personal Assistant Agent (PAA) is an AI assistant running as a text-based Windows desktop application to help users in scheduling and give reminders like a secretary. Users interact with the PAA conversationally. This project is a system of reasoning and a modular memory pipeline that makes any Large Language Model (LLM) behave like a personal assistant. The Language model is isolated inside a function to keep the system’s logic from interfering with the model’s weight. When the user enters their question or command into a basic graphical user interface (GUI) made with tkinter, the system fetches the user input and creates a prompt so the LLM can summarize a general idea of what the user wants. Then, it fetches relevant data from the vector memory system and/or searches the internet for relevant information. The system sends another query to the LLM with the supporting context and instructions to get a response to the command using a Python script to update or manage the requested application (Google Calendar, alarm, or notepad). In short, the central system maintains the context window, memory management, and a dynamic backend prompt to get the model responses and uses those responses. The system operates entirely on local hardware and demonstrates how a chatbot can be made from public language models. Potential users include researchers, businessmen, salesmen, and students interested in having a secretary that tailors the schedule for you and does trivial tasks like scheduling meetings, alarms, or taking notes. The end goal of the project is a prototype that can do the tasks mentioned above through an application on a laptop.
See coding folder here: Visit Personal-Assistant on GitHub
https://scholarship.depauw.edu/library_symposium/2025-2026/Spring2026/18
Comments
Completed as part of Professor Allana Johnson's CSC 498 Senior Project course.