New
Gear Up Your AI: Fine-Tuning LLMs
February 6, 2024

NextGen GPT AI Hackathon with Clarifai - Winners Announcement

Table of Contents:

Next Gen Winners Blog-1

To kick off 2024, we have partnered with Lablab.ai to host the NextGen GPT AI Hackathon with Clarifai. The purpose of the Hackathon is to nurture creativity in the development of AI-first products utilizing powerful models, including GPT-4 (Vision), Text to Speech, GPT-4 Turbo, and the Image Generation DALL·E 3 from the Clarifai community.  A total of 1,491 enthusiasts, forming 207 teams, have registered, resulting in an impressive 47 project submissions.

Building with GPT-4 Vision, GPT-4 Turbo, DALL.E 3 and other models from Clarifai Community

The challenge required participants to create a Generative AI project and showcase an innovative application using the GPT-4 Vision, GPT-4 Turbo, DALL.E 3, Text-to-Speech, and other diverse ranges of text, audio, and vision models from the Clarifai community.

47 impressive projects were created and submitted for judging. Here are some of the standout projects.

1st Place 🥇: Story Prep

Story Prep won the top prize. It is an app that helps you learn a language by reading your favourite author. You don't need a dictionary at hand to read the entire story. All you need to do is upload your story, and the application will convert the story into lessons that teach just what is necessary for you to enjoy that particular story.

It also allows you to generate vocabulary sheets and downloadable listening exercises based on any work of fiction. This material will teach you all the necessary words and grammar to jump into an uninterrupted reading experience.

Check out the project to learn more about the App.

Screenshot 2024-02-01 at 4.49.41 PM

2nd Place 🥈: AI Project Manager Assistant

AI Project Manager Assistant won second place with its aim to save time by keeping track of meeting notes. The team created a Slack bot that automatically records meeting transcripts and sends the summary in a specified Slack channel. The assistant has a /dailysummary command to get the summaries of all meetings during the day. 

Here's how AI Project Manager Assistant works:

1. Record a Meeting: As a first step, the assistant uses Fathom to record a Zoom meeting and sends the transcript as JSON to Zapier.

2. Convert JSON to .txt: Zapier converts JSON to .txt and sends as POST request to the server.

3. Summarize the transcript: The server calls the GPT-4-turbo model from Clarifai to summarize the .txt file.

4. Store the summary: The summarized .txt file is stored in the database.

5. Slackbot returns the summary: Finally, Slackbot returns the summarized meeting transcript upon typing /dailysummary.

Check out the project to learn more about AI Project Manager Assistants.

Screenshot 2024-02-01 at 5.00.14 PM

3rd Place 🥉: Recovery Pal

Recovery Pal, the third-place winner, is a mobile app designed for addiction recovery. The app supports various features, including:

1. Emotional Check-in to monitor and track a user's emotional state.

2. Custom Affirmations tailored to each user's needs and recovery progress, generated using the GPT-4 model from the Clarifai community.

3. Journaling for self-reflection.

4. Art Interpretation, where users can submit their art, and it will be interpreted by AI based on context; this is done using the GPT-4 Vision model and also uses the DALL·E 3 model to enhance the artistic expression.

5. Finally, Custom Meditations are created based on individual needs and preferences, which use the Text-to-speech model from the Clarifai community.

Check out the YouTube video below or project to learn more about Recovery Pal.

 

 

Special Mentions 🏆

Chatting with Historical Icons

Chatting with Historical Icons is an app designed to assist children with visual impairment, dyslexia, ADHD, auditory processing disorder, language disorders, and cognitive disabilities. The app's core features include:

1. Interactive Story Creation: By utilizing CrewAI, this application turns a child's ideas into personalized learning adventures, encouraging them to become story creators.

2. Rich Story Elements: Using models from the Clarifai community, the app generates diverse story elements (images, audio, text) to create an immersive, multi-sensory reading experience.

3. Dynamic Story Branching: The app integrates the object identification and the GPT-4 Vision models for real-time, interactive story alterations, enhancing engagement.

Food Complaint Resolution

The food complaint resolution application is an AI-based customer complaint system tailored to the food delivery industry. This app helps a food delivery company significantly reduce processing time and human effort in handling complaints. The application can swiftly process customer complaints, streamlining the resolution process.

It utilizes Food Item Recognition, GPT-4 Turbo, and GPT-4 Vision models from the Clarifai community.

EyeCue - Inspiring Recruiting Videos by AI

Traditional recruitment methods often neglect a substantial workforce segment: individuals not actively job hunting but open to a change. This widens the employee gap in many companies. EyeCue addresses this challenge by leveraging AI-generated videos to engage both active and passive candidates.

Final Words

We'd like to give a special thanks to Lablab.ai for partnering and hosting this hackathon. Congratulations to all of the hackers who came together to make it a success.

Narrowing down the winners was tough, with so many great projects submitted. Check out all the submitted concepts. prototypes and pitches from the Lablab.ai dashboard.

This was our first hackathon in 2024. It was great to see participants leveraging various text, vision, and audio models from the Clarifai community to build cool Generative AI projects. We were delighted and surprised by the submission number and quality. Can't wait to see what the hackers build for our next hackathon!

Want to stay in the loop for the next Clarifai Hackathon and developer events? Follow us on X and join the Discord Community to stay up to date!