top of page

GPT-Assistant:Platform to Build and Access Your Own Agents

OVERVIEW

GPT Assistant is a personalized, multi-functional assistant powered by Llama3, designed to streamline interactions by reducing repetitive inputs and organizing tasks by context. Currently, users often engage with language models in a single, linear chat thread, which can quickly become cluttered with various types of questions, requests, and prompts. This project aims to solve this problem by creating a dedicated tool with a user-friendly interface design that categorizes and optimizes user inputs across multiple specialized functions.

This project explores the potential of a modular GPT-based tool that enables users to type less, organize conversations, and focus more effectively on their goals. The interface will allow users to switch between modules quickly, each with predefined prompts and tones suited to different tasks.

This project is a prototype for a more robust and customizable assistant tool that could potentially support professionals, students, and creatives looking for a more organized AI interaction experience.

DOMAIN
SKILLS

Large Language Model

Web Development

JavaScript, HTML, CSS

System Prompt for Each Agent

​Enter instructions in the system prompt so users don't have to repeat instructions for similar tasks

Multiple Agents 

​Enter system instructions for each agent to categorize chats and avoid repeated typing.

Add New Agents 

​Users can add new agents or delete existing ones as needed.

Switch and Reorder Agents

Users can switch between different agents that store their chat history for a better experience.

They can also reorder agent tabs through drag-and-drop so they can have access to the most-frequently used agents more easily.

Let's Chat!

4th Floor, 370 Jay Street, Brooklyn, NY 11201 cy1503@nyu.edu  |  Tel: +1 914 447 2577

bottom of page