Because reality is overrated !
Problem Statement – Weave AI Magic with Groq
Deliver Personalized AI-Generated Content: Provide users with a daily dose of creative content — including stories, poems, news summaries, and puzzles — tailored to their interests and preferences using powerful LLMs.
Simplify Multilingual & Multimodal Interaction: Make content accessible and engaging through multilingual support and optional speech-to-text input, ensuring inclusivity across diverse user bases.
Enable Seamless Content Delivery & Engagement: Automate content scheduling and email delivery with a user-friendly interface, making daily engagement fun, effortless, and consistent.
X
- Velpula Vinay
- Sai Teja Gunda
- Chose this problem because content delivery is becoming monotonous and hard to personalize.
- Key challenges included integrating real-time news summarization, email scheduling, and cross-model support from Groq.
- Key challenges included integrating real-time news summarization, multilingual speech inputs, email scheduling, and cross-model support from Groq.
- Frontend: Streamlit
- Backend : Python, LangChain, Groq API, Firebase Authentication, Firebase Firestore
- Database: Firebase Firestore
- APIs: Groq API (for LLMs), Tavily Search API (for news)
- Hosting: Deployed using streamlit cloud
- ✅ Groq: Used to generate content using LLaMA3-70B and Mixtral for multilingual, stylistic, and fast inference.
- ✅ Tavily: Used for real-time web search and summarization of latest news.
Highlight the most important features of your project:
- ✅ Personalized story and poetry generator
- ✅ Dream interpretation and haiku creation
- ✅ Top 5 News Digest with real-time summarization
- ✅ User email subscription with instant or daily scheduler
- ✅ Groq model selection for customizable content tone and style
- Demo Video Link: https://youtu.be/pe6PzNMajEg?si=jnN8lfFcBuUg5O8V
- Deployment Link: https://dailyverse-hackhazards.streamlit.app/
- ✅ All members of the team completed the mandatory task - Followed at least 2 of our social channels and filled the form (Details in Participant Manual)
- ✅ All members of the team completed Bonus Task 1 - Sharing of Badges and filled the form (2 points) (Details in Participant Manual)
- ✅ All members of the team completed Bonus Task 2 - Signing up for Sprint.dev and filled the form (3 points) (Details in Participant Manual)
- Python 3.10+
- Firebase Project Setup (with Auth and Firestore)
- Groq API Key
- Tavily API Key
- .env file with:
- GROQ_API_KEY=your_key
- TAVILY_API_KEY=your_key
- FIREBASE_CONFIG=json_blob
# Clone the repo
git clone https://github.com/your-team/project-name
# Install dependencies
cd dailyverse
pip install -r requirements.txt
# Start development server
streamlit run dailyverse.py- 📈 User Dashboard: Add a personalized dashboard to track content history and update preferences
- 🔄 Recurring Delivery Scheduler (Cloud): Deploy a persistent scheduler using Docker on a cloud compute instance
- 🎙️ Voice Personalization: Integrate TTS (text-to-speech) for immersive auditory experiences
- 🌍 Localization: Expand to more regional languages with dynamic translations
- 🧠 Model Suggestions: Recommend best LLM models based on content type
- 📱 Mobile App Integration: Package into a native app with push notifications
- 🧠 Groq API: For ultra-fast inference and model flexibility (used for content generation)
- 🌐 Tavily API (via LangChain): To fetch and summarize top news from the web
- 🔤 LangChain Framework: For seamless chaining and LLM orchestration
- 📄 Streamlit: For building an interactive and intuitive frontend UI
- ☁️ Firebase: Used for authentication, database, and scheduled mail delivery
- 💌 smtp: For handling email delivery with a Gmail account
Building DailyVerse was more than just coding—it was a sprint of creativity, curiosity, and caffeine! From making LLMs generate quirky stories to sending them right to your inbox with Firebase magic, this journey was full of "aha!" moments and late-night bug hunts.
We explored real-time news summarization, model customization, and multilingual content delivery—all packed into a fun, user-first platform. Big cheers to the, hackathon organizers, and everyone who helped us debug that one stubborn issue (you know who you are).
Here’s to making AI not just smart, but soulful too. 🎉
