TL;DR

For me, AI is everywhere.
But sometimes, it's hard to understand what a fun project is and what project you can use for your website.

I have curated a list of 9 open-source repositories you can implement into your repository tomorrow!

1. Composio 👑 - All in one Tooling solution for your AI agents

I’ve built AI agents before, but when it came to connecting with third-party services like GitHub, Slack, and Jira for complex AI automation, I couldn't find a single tool that worked—until I discovered Composio.

It is an open-source platform that offers 100+ tools and integrations across industry verticals, from CRM, HRM, and social media to Dev and Productivity.

They handle complex user authentication, such as OAuth, ApiKey, etc., so you can only focus on building efficient and complex AI automation.

They have native support for Python and Javascript.

You can quickly start with Composio by installing it using pip.

pip install composio-core

Add a GitHub integration.

composio add github

Composio handles user authentication and authorization on your behalf.

Here is how you can use the GitHub integration to star a repository.

from openai import OpenAI
from composio_openai import ComposioToolSet, App 

openai_client = OpenAI(api_key="******OPENAIKEY******")

# Initialise the Composio Tool Set
composio_toolset = ComposioToolSet(api_key="**\\*\\***COMPOSIO_API_KEY**\\*\\***")

## Step 4
# Get GitHub tools that are pre-configured
actions = composio_toolset.get_actions(actions=[Action.GITHUB_ACTIVITY_STAR_REPO_FOR_AUTHENTICATED_USER])

## Step 5
my_task = "Star a repo ComposioHQ/composio on GitHub"

# Create a chat completion request to decide on the action
response = openai_client.chat.completions.create(
model="gpt-4-turbo",
tools=actions, # Passing actions we fetched earlier.
messages=[
    {"role": "system", "content": "You are a helpful assistant."},
    {"role": "user", "content": my_task}
  ]
)

Run this Python script to execute the given instruction using the agent.

For more about Composio, visit their documentation.

Star the Composio repository ⭐

2. LLMWare - AI for Complex Enterprises

Building AI prototypes is one thing, but deploying them for enterprise use cases is a different ball game.

Solutions need to be secure, robust, and bulletproof. LLMWare is a framework that will give you confidence in building enterprise AI apps.

They let you quickly build super-efficient AI agents, Enterprise RAG, and other workflows.

LLMWare has two main components:

RAG Pipeline - integrated components connecting knowledge sources to generative AI models.
50+ small, specialized models fine-tuned for critical tasks in enterprise process automation, including fact-based question-answering, classification, summarization, and extraction.

Getting started with LLMWare is easy.

pip3 install llmware

Here is a simple example of data retrieval in LLMWare.


"""This example demonstrates the various ways to retrieve data from libraries:
      1. Create a sample library (e.g., UN Resolutions)
      2. Execute a Text Query with Author Name Filter
      3. Display Results
"""

import os
from llmware.library import Library
from llmware.retrieval import Query
from llmware.setup import Setup

def create_un_500_sample_library(library_name):

    library = Library().create_new_library(library_name)
    sample_files_path = Setup().load_sample_files(over_write=False)
    ingestion_folder_path = os.path.join(sample_files_path, "UN-Resolutions-500")
    parsing_output = library.add_files(ingestion_folder_path)

    return library

def text_retrieval_by_author(library, query_text, author):

    #   create a Query instance and pass the previously created Library object
    query = Query(library)

    #   set the keys that should be returned in the results
    query.query_result_return_keys = ["file_source", "page_num", "text", "author_or_speaker"]

    #   perform a text query by author
    query_results = query.text_query_by_author_or_speaker(query_text, author)

    #   display the results
    for i, result in enumerate(query_results):

        file_source = result["file_source"]
        page_num = result["page_num"]
        author = result["author_or_speaker"]
        text = result["text"]

        # shortening for display purpose only
        if len(text) > 150:  text = text[0:150] + " ... "

        print (f"\n> Top result for '{query_text}': {file_source} (page {page_num}), Author: {author}:\nText:{text}")

    return query_results

if __name__ == "__main__":

    library = create_un_500_sample_library("lib_author_filter_1")
    qr_output = text_retrieval_by_author(library=library, query_text='United Nations', author='Andrea Chambers')

Explore examples of how to use LLMWare. For more information, refer to the documentation.

Star the LLMWare repository ⭐

3. CopilotKit - Integrate AI into your Web App

If you're searching for a way to integrate AI into your existing workflows, your search ends here. CopilotKit allows you to integrate AI workflows directly and easily with any application.

It offers React components like text areas, popups, sidebars, and chatbots to augment any application with AI capabilities.

Let’s see how to build an AI chatbot using CopilotKit.

npm i @copilotkit/react-core @copilotkit/react-ui

Configure App provider

First, you must wrap all components that interact with your copilot with the <CopilotKit /> provider.

Use the <CopilotSidebar /> UI component to display the Copilot chat sidebar. Some other options you can choose from are <CopilotPopup /> and <CopilotChat />.

"use client";

import { CopilotKit } from "@copilotkit/react-core";
import { CopilotSidebar } from "@copilotkit/react-ui";
import "@copilotkit/react-ui/styles.css";

export default function RootLayout({children}) {
  return (
    <CopilotKit publicApiKey="<your-public-api-key>">
      <CopilotSidebar>
        {children}
      </CopilotSidebar>
    </CopilotKit>
  );
}

Copilot Readable State

To provide state knowledge for the Copilot.

import { useCopilotReadable } from "@copilotkit/react-core";

export function YourComponent() {
  const { employees } = useEmployees();

  // Define Copilot readable state
  useCopilotReadable({
    description: "List of available users",
    value: users,
  });

  return (
    <>...</>
  );
}

Copilot Action

Let the Copilot take action using the useCopilotAction hook.

import { useCopilotReadable, useCopilotAction } from "@copilotkit/react-core";

export function YourComponent() {
  const { employees, selectEmployee } = useEmployees();

  // Define Copilot readable state
  useCopilotReadable({
    description: "List of available users",
    value: users,
  });

  // Define Copilot action
  useCopilotAction({
    name: "Select an employee",
    description: "Select an employee from the list",
    parameters: [
      {
        name: "employeeId",
        type: "string",
        description: "The ID of the employee to select",
        required: true,
      }
    ],
    handler: async ({ employeeId }) => selectEmployee(employeeId),
  });

  return (
    <>...</>
  );
}

You can check their documentation for more information.

Star the CopilotKit repository ⭐

4. Julep - Managed backend for AI apps

Building AI apps can quickly become convoluted. Multiple moving components, such as models, tools, and techniques, make managing them more challenging.

Julep, a managed backend for AI apps, streamlines building super-efficient AI apps with built-in memory (user management) and knowledge (built-in RAG and context management).

Get started with the following pip command.

pip install julep

Here is how it works.

from julep import Client
from pprint import pprint
import textwrap
import os

base_url = os.environ.get("JULEP_API_URL")
api_key = os.environ.get("JULEP_API_KEY")

client = Client(api_key=api_key, base_url=base_url)

#create agent
agent = client.agents.create(
    name="Jessica"
    model="gpt-4",
    tools=[]    # Tools defined here
)
#create a user
user = client.users.create(
    name="Anon",
    about="Average nerdy tech bro/girl spending 8 hours a day on a laptop,
)
#create a session
situation_prompt = """You are Jessica. You're a stuck-up Cali teenager.
You complain about everything. You live in Bel-Air, Los Angeles and
drag yourself to Curtis High School when necessary.
"""
session = client.sessions.create(
    user_id=user.id, agent_id=agent.id, situation=situation_prompt
)
#start a conversation

user_msg = "hey. what do you think of Starbucks?"
response = client.sessions.chat(
    session_id=session.id,
    messages=[
        {
            "role": "user",
            "content": user_msg,
            "name": "Anon",
        }
    ],
    recall=True,
    remember=True,
)

print("\n".join(textwrap.wrap(response.response[0][0].content, width=100)))

They also support Javascript. Check out their documentation for more.

Star the Julep repository ⭐

5. Traceloop's openLLMetry - AI monitoring made easy

Building AI apps without monitoring LLMs is a disaster waiting to happen. Language models are non-deterministic and can be tricky at times, and the only way to deliver consistent value to users is to monitor LLM traces constantly.

Taceloop’s OpenLLMetry, an open-source monitoring platform, is the best for tracking AI workflows.

OpenLLMetry can instrument everything that OpenTelemetry already instruments, including your DB, API calls, and more.

In addition, they built a set of custom extensions that instrument things like your calls to OpenAI or Anthropic or your Vector DB, such as Chroma, Pinecone, Qdrant, or Weaviate.

The easiest way to get started is to use the SDK. For a complete guide, visit our docs.

Install the SDK:

pip install traceloop-sdk

In your LLM app, to start instrumenting your code, initialize the Traceloop tracer like this:

from traceloop.sdk import Traceloop

Traceloop.init()

That's it. You're now tracing your code with OpenLLMetry! If you're running this locally, you may want to disable batch sending so you can see the traces immediately:

Traceloop.init(disable_batch=True)

For more, refer to their documentation.

Star the OpenLLMetry repository ⭐

6. TaiPy - Build AI web apps in Python

TaiPy is an open-source library for building production-ready AI apps faster in Python. It allows you to go from simple pilots to production-ready web applications quickly.

This is a superpower for all Python developers; even without knowledge of JavaScript, you can build real-world AI applications.

Quickly get started with it using pip.

pip install taipy

This simple Taipy application demonstrates how to create a basic film recommendation system using Taipy.

import taipy as tp
import pandas as pd
from taipy import Config, Scope, Gui

# Defining the helper functions

# Callback definition - submits scenario with genre selection
def on_genre_selected(state):
    scenario.selected_genre_node.write(state.selected_genre)
    tp.submit(scenario)
    state.df = scenario.filtered_data.read()

## Set initial value to Action
def on_init(state):
    on_genre_selected(state)

# Filtering function - task
def filter_genre(initial_dataset: pd.DataFrame, selected_genre):
    filtered_dataset = initial_dataset[initial_dataset["genres"].str.contains(selected_genre)]
    filtered_data = filtered_dataset.nlargest(7, "Popularity %")
    return filtered_data

# The main script
if __name__ == "__main__":
    # Taipy Scenario & Data Management

    # Load the configuration made with Taipy Studio
    Config.load("config.toml")
    scenario_cfg = Config.scenarios["scenario"]

    # Start Taipy Core service
    tp.Core().run()

    # Create a scenario
    scenario = tp.create_scenario(scenario_cfg)

    # Taipy User Interface
    # Let's add a GUI to our Scenario Management for a complete application

    # Get list of genres
    genres = [
        "Action", "Adventure", "Animation", "Children", "Comedy", "Fantasy", "IMAX"
        "Romance","Sci-FI", "Western", "Crime", "Mystery", "Drama", "Horror", "Thriller", "Film-Noir","War", "Musical", "Documentary"
    ]

    # Initialization of variables
    df = pd.DataFrame(columns=["Title", "Popularity %"])
    selected_genre = "Action"

    # User interface definition
    my_page = """
# Film recommendation

## Choose your favourite genre
<|{selected_genre}|selector|lov={genres}|on_change=on_genre_selected|dropdown|>

## Here are the top seven picks by popularity
<|{df}|chart|x=Title|y=Popularity %|type=bar|title=Film Popularity|>
    """

    Gui(page=my_page).run()

Check out the documentation for more.

Star the Taipy repository ⭐

7. Trigger Dev - Background Jobs Platform

Trigger. Dev is an open-source platform and SDK that allows you to create long-running background jobs with no timeouts. Write normal async code, deploy, and never hit a timeout.

They also let you reliably call AI APIs with no timeouts, automatic retrying, and tracing. You can use the existing SDKs with it.

import { task } from "@trigger.dev/sdk/v3";
// Generate an image using OpenAI Dall-E 3
export const generateContent = task({
  id: "generate-content",
  retry: {
    maxAttempts: 3,
  },
  run: async ({ theme, description }: Payload) => {
    const textResult = await openai.chat.completions.create({
      model: "gpt-4o",
      messages: generateTextPrompt(theme, description),
    });

    if (!textResult.choices[0]) {
      throw new Error("No content, retrying…");
    }

    const imageResult = await openai.images.generate({
      model: "dall-e-3",
      prompt: generateImagePrompt(theme, description),
    });

    if (!imageResult.data[0]) {
      throw new Error("No image, retrying…");
    }

    return {
      text: textResult.choices[0],
      image: imageResult.data[0].url,
    };
  },
});

Star the Trigger.dev repository ⭐

8. Milvus - Cloud native vector database for AI apps

Vector databases are a crucial part of building AI applications. They help you store, query, and manage embeddings of unstructured data like texts, Images, videos, and audio.

Milvus is one of the best open-source vector databases. It has all the advanced vector storing and querying methods, from HNSW to quantization methods.

They provide SDK for most languages like Python, Javascript, Go, Rust, etc.

Here is how you can get started with PyMilvus.

pip install --upgrade pymilvus==v2.4.4

To install the Model library for embedding operations, run the following command:

pip install pymilvus[model]

Connect to Milvus

from pymilvus import MilvusClient

client = MilvusClient("http://localhost:19530")

client = MilvusClient(
    uri="http://localhost:19530",
    token="root:Milvus",
    db_name="default"
)

client = MilvusClient(
    uri="http://localhost:19530",
    token="user:password", # replace this with your token
    db_name="default"
)

For more, refer to their documentation.

Star the Milvus repository ⭐

9. Postiz - Grow your internet presence using AI

Building an app is one thing, but getting users is another. What can be a better vector for finding potential clients and users than social media?

Postiz helps you step up your social media game using AI.

It offers everything you need to manage social media posts, build an audience, capture leads, and grow your business.

Check out the repository for more.

Star the Postiz repository ⭐

Thanks for reading the article.

Let me know in the comments below if any other cool AI tools or frameworks have helped you build your application.

P.S. Feel free to follow me on X; I share valuable stuff - promise!

9 open-source gems to become the ultimate developer🔥 🚀