arammyy94

No more AI toy projects that end with a single API call!
Based on my experience designing and deploying a full-stack AI chatbot, I will guide you through the entire process of building an AI chatbot that "works as a real service"—covering streaming, tool calling, memory, multi-modal capabilities, and Human-in-the-Loop.

kaburi

Building a Full-Stack AI ChatBot (React 19, NestJS, LangChain.Graph v1.+)

LangChain

langchain

fullstack

human-in-the-loop

llm-streaming

ai-chatbot

Kaburi

입문

초급

중급이상

리액트

React

포스트그레SQL

PostgreSQL

네스트JS

NestJS

랭체인

랭그래프

LangGraph

Intro

Frontend Project Initial Setup

Creating a chatbot UI screen

Backend (NestJS) Project Initial Setup

LangChain Integration and Chat Streaming

Frontend Refactoring

LangChain(v1.0) Agent — Tool Calling and Streaming

Displaying the tool execution process through the UI

Multimodal — Image and PDF Processing

Remembering Conversations — LangGraph Short-term Memory

Human in the Loop — Users determine the actions of the AI

Separate Chat Rooms — Independent Memory by Chat ID

Build a PostgreSQL database

Chat Data Persistence — Prisma + PostgreSQL