Event ended

AI Agents Hackathon Summary

AI Agents Hackathon image

Hackathon Overview

Our AI hackathon brought together a diverse group of participants, who collaborated to develop a variety of impressive projects based on:

589

Participants

84

Teams

15

AI Applications

Winners and Finalists

Speakers, Mentors, and Organizers

Paulo Almeida profile picture
mentor

Paulo Almeida

co-founder of Stunning Green

This event has now ended, but you can still register for upcoming events on lablab.ai. We look forward to seeing you at the next one!

Checkout Upcoming Events →

Submitted Concepts, Prototypes and Pitches

Submissions from the teams participating in the AI Agents Hackathon event and making it to the end 👊

Help to spread the word and share these amazing projects!

Let them live - Auto-GPT Virtual Influencer

"Let Them Live" employs these capabilities to create Virtual Influencers that can autonomously perform any action a real influencer can do on social media platforms. At present, it operates on Instagram, where it can autonomously log into your account, initiate and maintain conversations, follow users based on specific topics, and even generate consistent and personalized images using the Dreambooth model. All these actions are facilitated by the instagrapi Python library and the FAISS vector store for efficient retrieval of past messages. But the potential of "Let Them Live" extends beyond Instagram. We are currently working on integrating Twitter into our platform, which will allow our Auto-GPT Virtual Influencers to autonomously engage in interactions on Twitter, further expanding their reach.

Let them live

LangChainOpenAIAuto-GPT

IntelliFind

IntelliFind is a revolutionary tool designed to harness the power of Intelligent Document Processing (IDP) and streamline the document management process. Our project utilizes the latest advancements in Language Model Embeddings, specifically Longformer Language Model (LLM) embeddings, to effectively index vast collections of documents and store them in a vector database. Our tool will take unstructured document data, process it using our LLM embeddings, and convert it into a searchable, structured format. This innovative approach allows users to easily search through massive amounts of data with high accuracy and speed, facilitating fast information retrieval and improved workflow efficiency. Additionally, IntelliFind goes beyond mere search capabilities. We've integrated a unique chat feature that provides contextual responses based on the indexed documents. The interactive chat employs our sophisticated AI model to maintain the context of the conversation, assisting users in exploring and extracting detailed insights from their document repositories. From corporate data management to academic research, IntelliFind empowers users to handle information overload, efficiently navigate through complex document structures, and extract the most value from their data assets. Whether you are searching for a specific document or probing for intricate patterns in vast amounts of data, IntelliFind ensures a user-friendly and seamless experience, making information retrieval a breeze.

Deep Index

OpenAIChromaLangChain
Streamlit
application badge

MindSpeak - Visualizing Mental Health Support

MindSpeak is a groundbreaking project that leverages cutting-edge technologies to revolutionize mental health support. Mental health disorders are prevalent in our society, but due to stigmatization and lack of accessible information, many individuals face challenges in seeking help and finding accurate resources. The project combines the power of artificial intelligence, advanced embedding techniques, and immersive multimedia to offer an engaging and interactive platform for mental health support. The process begins with Chroma, an innovative tool that converts uploaded PDF files into vector representations. By employing advanced embedding techniques, Chroma ensures that the information is accurately captured and transformed into a format suitable for further processing. To enhance the quality of vectorization and improve the overall representation, Cohere comes into play. Cohere facilitates the embedding process, utilizing sophisticated algorithms to refine and enhance the vectorized data. This step ensures that the generated vectors are of high quality and accurately capture the nuances of the original content. One of the key features of MindSpeak is Stable Diffusion, a technology that enables the generation of coherent and visually appealing images based on the text generated by the model. By analyzing the textual information, Stable Diffusion generates images that align with and enhance the provided content. To further enhance the user experience and accessibility, MindSpeak incorporates Elevenlabs, a powerful tool that converts text into speech. This feature allows the generated content to be conveyed audibly, adding an immersive audio component to the multimedia animations.

Cookies

ChromaCohereCohere EmbedStable Diffusion
Streamlit
application badge

SmartApply

Our project aims to revolutionize the resume creation process by leveraging the power of AI Agents and web scraping. We understand that crafting a compelling resume can be a time-consuming and challenging task, and our tool is designed to simplify this process. By harnessing the capabilities of artificial intelligence and utilizing web scraping techniques, we have developed a custom resume creator that helps individuals create personalized resumes tailored to their unique skills, experiences, and career goals. Our AI Agents analyze user input, extract relevant information from online sources, and generate dynamic and industry-relevant content for each resume section. The result is a neatly formatted resume that stands out to potential employers and increases the chances of securing job opportunities. With user-friendly features, real-time editing options, and the ability to export and share resumes, our tool empowers users to confidently present their qualifications and achievements in a professional manner. Whether you're a seasoned professional or a career changer, our custom resume creator is here to support you in showcasing your potential and landing your dream job.

SmartApply

gpt4allChatGPTLangChainCAMELOpenAIGenerative AgentsGPT-3.5Auto-GPTGPT-4

Chatty Shoes

Chatty Shoes is designed with a highly responsive, scalable and robust backend, leveraging state-of-the-art technology to facilitate user-friendly online shoe shopping via natural language interactions. The backbone of the system is a serverless architecture based on Cloud Functions. This decentralised structure comprises three primary functions: Session Creation, Message Management and Information Retrieval. Session Creation: This function is responsible for initiating conversations with customers. Every conversation represents a unique session, providing the framework for interactive and dynamic dialogues. The data from these sessions, including the conversation history, is securely stored in Firestore, ensuring a persistent and seamless user experience across different sessions. Message Management: Powering the heart of the conversation is a GPT-3.5-turbo, a language model renowned for its ability to understand and generate human-like text. This feature enables Chatty Shoes to handle Frequently Asked Questions (FAQs) within the context of the conversation, thereby giving users instant responses to their queries. Additionally, it provides the AI agent with the ability to perform information retrieval operations on products when the system deems it necessary, making it proactive and more engaging. Information Retrieval: The last key component of the backend is the Semantic Search function, which operates over Pinecone's vectorial database. This feature enhances Chatty Shoes' product recommendation abilities by retrieving product data based on the contextual understanding of user inputs rather than merely keyword matching. This means the AI can effectively understand and respond to nuanced customer preferences, thus improving their overall shopping experience. Together, these elements synergise to create a powerful backend that underpins Chatty Shoes' mission: to revolutionize online shoe shopping by facilitating better, deeper and more intuitive conversations.

Chatty shoes

OpenAIChatGPTLangChainGPT-3.5
Streamlit
application badge

Multiagent

This is a repository that was made for a Hackathon orginized by Lablab.AI. The challenge was to create different types of agents that will carry our several tasks. Use the power of LLMs with LangChain and OpenAI to scan through your documents. Find information and insight's with lightning speed. 🚀 Create new content with the support of state of the art language models and and voice command your way through your documents. 🎙️""") st.write("We wills how you 5 different agents that we build\n" "1. **AssemblyAI Agent**\n" "2. **PandasAI Agent**\n" "3. **Presentation Agent**\n" "4. **README Agent**\n" "5. **Webscraping generator Agent**\n

RONG

OpenAIWhisperLangChain

Auto Claude 100k

Claude provides an API that accepts up to 100k tokens. AutoGPT faces challenges with memory and retrieval, which you can learn about at https://docs.agpt.co/challenges. Therefore, we can use the Claude API, which accommodates 100k tokens, to run AutoGPT and avoid these memory and retrieval challenges. So, I've modified the AutoGPT source code. I have replaced the chat API with Claude and disabled memory. Another benefit of using AutoClaude is that it's free! The Claude API does have a rate limit for each request, but there's no limit on usage frequency. Therefore, we can run the Claude API continuously.

Life unlimited

Anthropic ClaudeAuto-GPT

Trading-Agent-

A trading agent AI is an artificial intelligence system that uses computational intelligence methods such as machine learning and deep reinforcement learning to automatically discover, implement, and fine-tune strategies for autonomous adaptive automated trading in financial markets This project implements a Stock Trading Bot, trained using Deep Reinforcement Learning, specifically Deep Q-learning. Implementation is kept simple and as close as possible to the algorithm discussed in the paper, for learning purposes. Generally, Reinforcement Learning is a family of machine learning techniques that allow us to create intelligent agents that learn from the environment by interacting with it, as they learn an optimal policy by trial and error. This is especially useful in many real world tasks where supervised learning might not be the best approach due to various reasons like nature of task itself, lack of appropriate labelled data, etc. The important idea here is that this technique can be applied to any real world task that can be described loosely as a Markovian process. This work uses a Model-free Reinforcement Learning technique called Deep Q-Learning (neural variant of Q-Learning). At any given time (episode), an agent abserves it's current state (n-day window stock price representation), selects and performs an action (buy/sell/hold), observes a subsequent state, receives some reward signal (difference in portfolio position) and lastly adjusts it's parameters based on the gradient of the loss computed. There have been several improvements to the Q-learning algorithm over the years, and a few have been implemented in this project: Vanilla DQN DQN with fixed target distribution Double DQN Prioritized Experience Replay Dueling Network Architectures Trained on GOOG 2010-17 stock data, tested on 2019 with a profit of $1141.45 (validated on 2018 with profit of $863.41):

Cyber World

Reinforcement LearningprivateGPTgpt4allChatGPT

ClassA-GPT

Setting up a class action lawsuit is not easy. Lawyers struggle to streamline and communicate so many people. Participants might not know that they are eligible in the first place. This makes class actions lawsuits a long and complex process. We propose an AI moderator that has easily checks people's current situation, eligibility to join a lawsuit, onboards them and acts as a single point of contact for case's progress. Empowering the participant to understands the situation quicker because the AI guides people through understanding the legal case and providing relevant information. Empowering the lawyer to focus on strategic decision making because all the information from case participants is available. Video: https://www.youtube.com/watch?v=2KHAxcJ3c90 Demo: https://classa.herokuapp.com/

Hackaheads

GPT-4
Streamlit
application badge

HyperCompare AI

HyperCompare AI, the game-changing hyperintelligent solution that revolutionizes the way you compare products. With lightning speed and unmatched accuracy, this cutting-edge AI system scours the vast expanse of the internet, tirelessly gathering data from e-commerce websites, expert reviews, and customer feedback platforms. It analyzes and synthesizes this information into comprehensive and unbiased product comparisons, empowering you with the insights you need to make confident purchasing decisions. Say farewell to the endless hours of manual research and hello to the future of smart shopping. HyperCompare AI: your trusted companion in navigating the sea of options, unlocking a world of informed choices at your fingertips. Experience the transformative power of HyperCompare AI and embark on a new era of shopping excellence.

TheBanggog

OpenAILangChainText Generation Web UIGenerative Agents

Remote-how AI

Remote-how AI is a game-changing Slack companion designed to optimize remote work. Since 2017, Remote-how has been a trusted partner for over 3,000 companies worldwide, and now we're taking it a step further with AI. Our tool assesses your team's remote work performance, generating a unique Remote Work Score. Based on this score, it provides personalized improvement plans, pinpointing areas for enhancement. It also gives you access to a comprehensive Remote-First Playbook, a living document that evolves with your team's needs. Experience the future of remote work with Remote-how AI, your personal remote work strategist.

AI-AI-Boys

OpenAILangChainStable Diffusion
Streamlit
application badge

LangX Learning languages made easy

LangX's AI agent that can easily teach you a language quite fast we also do it for free thus increasing accessibility and affordability. We aim to make learning languages easy in remote countries for professional/personal developmental purposes this also increases opportunities for proffesionals as many companies are looking people who know many languages. It also helps you save time by just sitting and using the website instead of going to the teacher's house or attending classes and preventing boredom for students not wanting to attend classes thus increasing linguistic interest in children.

LangX

LangChainOpenAIReinforcement Learning

All AI combined in one place

Why more than one AI? My idea is that we collect all artificial intelligence data and services such as GPT chat, images, text, programming, design, sound and everything that can be useful. Example: If a programmer does not have an idea, then he will not do anything, and also the owner of the idea, but when you combine them, they produce creativity. Likewise, this is my idea, when the data of each of the realities of artificial intelligence cooperate in one place. , In addition, it will increase people's demand for one site instead of ten sites. In detail: We will collect data from artificial intelligence (for example, from GBT chat and creating images) when we collect them and add to them that they help in creating everything that is required of you, as someone asks him I want a picture in which there will be, for example, a schedule for my daily life (and here is the second stage of the idea) he will ask The Ai who asked for the photo will tell him what are your goals and what are your dates (in addition, he can connect it to the calendar on his phone to know what his dates are). Then, when the Ai finishes asking him, he will produce the photo exactly as he requested it. I know that you will say that it can also be created in an image site only, it is true, but when a person is talking to the Ai in words and exchanging conversation with him, GBT chat will advise him and tell him suggestions, show him examples and and and more possibilities, imagine this only from two sites, if we add more The services will be as if an expert person will help you create what you want perfectly This is only the general idea, and that in the details of the site, services covering all categories will be provided For example, for those who do not see, and for those who do not have hands, we will provide them with how to deal with the site.

All in one

Streamlit
application badge

IdeaStorm

Ideastorm is an AI-powered chat platform designed to spark creativity and generate fresh ideas. By configuring two distinct generative AI agents with unique names, descriptions, and tool access, users can initiate engaging discussions on any given topic. Each agent's distinct experiences, characteristics, and tools lead to a dynamic conversation that offers new perspectives and insights. This versatile platform is perfect for marketing teams seeking to brainstorm innovative campaigns, as well as design and development teams looking for fresh concepts. Ideastorm's unique approach to idea generation sets it apart from traditional brainstorming methods, as it harnesses the power of AI to simulate thought-provoking conversations between two AI agents. Unlock your team's potential and explore uncharted creative territory with Ideastorm, the ultimate tool for brainstorming, idea generation, and gaining new perspectives through AI-driven discussions.

IdeaStorm

OpenAILangChainGenerative AgentsOpenAI gym

College CareerSage

During our peer-mentoring process, we discovered that college students often struggle with career development, overwhelmed by academic obligations and extracurricular activities. Many find it difficult to determine if their efforts are aligned with their career objectives and secure mentors who understand their experiences and provide tailored guidance. The abundance of information available can be overwhelming, leading to missed deadlines for scholarships and career opportunities. Unfortunately, there is a shortage of mentors available to meet the 90% of students who require personalized mentorship. To address these challenges, we created CareerSage, an AI-powered solution for college students. CareerSage utilizes web scraping to build a private dataset of university information, enabling personalized career development based on students' experiences and preferences. With the power of GPT models, CareerSage offers timely, reliable information, a tailored step-by-step plan, an AI mentor bot, and effortless personal experience management. Students benefit from up-to-date information on scholarships and career opportunities, ensuring they stay informed. They receive a personalized plan that takes into account their unique interests, skills, and aspirations. The AI mentor bot acts as a virtual guide, providing relevant advice and suggestions. Additionally, CareerSage simplifies the management of students' personal experiences, helping them track accomplishments and present them effectively. CareerSage empowers college students by bridging the gap between their aspirations and the guidance they need. With a comprehensive and tailored approach, students can make informed decisions and achieve success in their chosen career paths.

XSage

LangChainChatGPTOpenAI