Labs for the "Build an agentic LLM assistant on AWS" workshop. A step by step agentic llm assistant development workshop using serverless three-tier architecture.

Supercharged AI Dev Tools is an open, expertly curated directory for the latest and most powerful AI coding assistants, LLMs, autocompletion engines, and automation agents. Discover tools that accelerate software engineering, enhance code quality, automate reviews and testing, and let you build applications faster than ever—all powered by AI.

Agent Aide, an LLM-powered digital assistant to build teams and answer questions about VALORANT Esports players for the VCT x AWS Hackathon

using llm agents to build workflows and handle them dynamically to be an assistant to the user

Build an Catering Service AI agent assistant using Gemini LLM model and Smolagent HuggingFace Library

No description available

No description available

This project showcases how to combine Retool, LLM-powered assistants, and structured databases to build an intelligent, agentic healthcare assistant. The AI system dynamically interacts with users, interprets their intent, and autonomously triggers database operations—bridging the gap between conversational AI and real-world action.

This repository contains an implementation of an intelligent currency-conversion assistant built using the Google Agent Development Kit (ADK). The notebook demonstrates how to build a structured LLM agent capable of performing precise currency conversions, handling financial instructions, and executing requests with reasoning steps.

An Intelligent Document Q&A Agent using RAG with DeepSeek LLM & Ollama, powered by Streamlit. DocReader AI Agent allows users to upload PDF documents and ask context-aware questions using a RAG pipeline. It combines the power of LangChain, DeepSeek LLM via Ollama, and Streamlit to build a lightweight, local-first document assistant.

A minimal, educational implementation of an agentic coding assistant inspired by tools like Claude Code, Cursor, and OpenCode, built using Google’s Gemini API. This project explores how simple it is to build a custom AI agent using an LLM. It uses gemini-2.5-flash and is intended for learning and experimentation, not production use.

Build an AI Operations Assistant that accepts a natural-language task, plans steps, calls tools (APIs), and returns a final structured answer. The system must run locally and demonstrate agent-based reasoning, LLM usage, and real API integrations.

ChatGPT said: This assignment demonstrates how to build a simple AI assistant using the OpenAI Agents SDK and integrate it with Chainlit to provide an interactive web-based chat interface. It shows how the agent leverages LLM capabilities to respond to user queries in real time.

This project builds an intelligent to-do assistant on Coze using AI agents and LLMs. It supports fuzzy Chinese input, structured task extraction, auto-categorization, priority reminders, status tracking, and completion visualization. Work includes system design, development, testing, optimization, and thesis.

This project builds an intelligent university information assistant using LangChain’s MCP Agent and FastAPI tools. It retrieves official university websites, extracts application deadlines, and returns CS rankings. The system integrates OpenAI LLMs, BeautifulSoup scraping, and CollegeScorecard API. Ideal for demonstrating real-world tool -integ

E-commerce platforms receive large volumes of queries about orders, returns, payments, and policies. Human agents alone are slow and costly to scale. You will build an AI assistant that can resolve common issues autonomously using LLMs, RAG, and simple tool calls, with clear fallbacks when it cannot help.

Dify (Dify (Open-Source No-Code AI Platform) Dify is an open-source, no-code / low-code AI application platform that lets you build LLM apps, chatbots, and AI agents using a visual interface—with little to no programming. --- 🔑 What Dify Does Build AI apps visually (chatbots, Q&A bots, assistants) Create workflows (prompt chains, tool Licen