Code Vidya

Best Use of the Google Gemini API

It’s time to push the boundaries of what's possible with AI using Google Gemini. Check out the Gemini API to build AI-powered apps that make your friends say WHOA. So, what can Gemini do for your hackathon project? * Understand language like a human and build a chatbot that gives personalized advice * Analyze info like a supercomputer and create an app that summarizes complex research papers * Generate creative content like code, scripts, music, and more Think of the possibilities… what will you build with the Google Gemini API today?

PS 1 — Smart Anchor & Stage Flow Management System. Problem Statement - College events such as hackathons, workshops, seminars, competitions, and cultural programs involve multiple speakers, activities, and strict timelines. During the event, anchors and organizers often have to manage introductions, transitions, timings, audience engagement, schedule changes, and unexpected announcements at the same time. Build a real-time event management platform that helps organizers plan the event and supports anchors during live execution with relevant information, context-aware scripts, and quick assistance when the schedule changes

Expected Deliverable • Event and agenda management • Speaker and guest management • AI-generated opening, introduction, transition, and closing scripts • Live dashboard showing the current and upcoming activities • Dynamic agenda updates • Support for delays and unexpected announcements • At least one meaningful AI-powered application workflow Judging Parameters • Relevance and quality of generated scripts • Effective use of event and speaker information • Usability of the live stage dashboard • Handling of last-minute schedule changes • Quality of AI integration • Overall functionality of the MVP Technology Parameters AI Models: Gemini, ChatGPT, Claude, Llama, Mistral, etc. APIs: Gemini API, OpenAI API, Anthropic API, other suitable APIs, etc. Frontend: React, Next.js, HTML/CSS/JavaScript, Tailwind CSS, Bootstrap, etc. Backend: Node.js, Python, Java, PHP, etc. Database: MongoDB, PostgreSQL, MySQL, SQLite, Firebase, JSON, etc. Other: WebSockets, voice/Live APIs, authentication, notifications, data visualization, etc.

PS 2 — Dynamic Certificate Generation & Verification Portal. Problem Statement - College events can involve hundreds or even thousands of participants. Managing certificates manually can lead to incorrect names, duplicate records, missing information, formatting issues, and difficulties in verifying whether a certificate is genuine. Build an end-to-end platform that simplifies the complete certificate process—from checking participant data and generating certificates in bulk to providing a simple and reliable verification system.

Expected Deliverable • Event and certificate template management • Bulk participant upload through CSV • AI-assisted data cleaning and validation • Human review before final certificate generation • Bulk certificate generation • Unique certificate IDs • QR-based certificate verification • Public verification page Judging Parameters • Accuracy of participant-data validation • Quality and consistency of generated certificates • Handling of duplicate, missing, or incorrect data • Reliability of the verification system • Efficiency of bulk certificate generation • Overall usability and functionality Technology Parameters AI Models: Gemini, ChatGPT, Claude, Llama, Mistral, etc. APIs: Gemini API, OpenAI API, other suitable APIs, etc. Frontend: React, Next.js, HTML/CSS/JavaScript, etc. Backend: Node.js, Python, Java, PHP, etc. Database: MongoDB, PostgreSQL, MySQL, SQLite, Firebase, JSON, etc. Other: QR generation/scanning, PDF generation, OCR, CSV processing, authentication, etc.

PS 3 — ClubOps AI. Problem Statement - College clubs often manage events through WhatsApp groups, spreadsheets, documents, meeting notes, and personal task lists. As an event grows, it becomes difficult to keep track of responsibilities, deadlines, pending tasks, dependencies, and potential risks. Build a centralized platform that helps clubs plan and execute events by bringing tasks, volunteers, meetings, deadlines, documents, and risks together in one place.

Expected Deliverable • AI-assisted event planning • Task and volunteer management • Meeting notes or transcript processing • Automatic identification of action items, owners, and deadlines • Risk identification and explanation • Club document and knowledge repository • AI-assisted announcements and communication • AI-assisted workflows that can perform actual application actions Judging Parameters • Usefulness and accuracy of event planning • Accuracy of task, owner, and deadline extraction • Practicality of risk detection • Effectiveness of the club knowledge base • Quality of task and volunteer management • Quality of AI integration Technology Parameters AI Models: Gemini, ChatGPT, Claude, Llama, Mistral, etc. APIs: Gemini API, OpenAI API, other suitable APIs, etc. Frontend: React, Next.js, HTML/CSS/JavaScript, Tailwind CSS, etc. Backend: Node.js, Python, Java, PHP, etc. Database: MongoDB, PostgreSQL, MySQL, SQLite, Firebase, JSON, etc. Other: RAG, document processing, authentication, notifications, dashboards, etc.

PS 4 — CampusContent AI. Problem Statement - Promoting a college event usually means creating separate content for Instagram, LinkedIn, WhatsApp, email, registration pages, posters, and websites. Creating everything separately takes time and can easily lead to inconsistent dates, timings, messaging, or tone. Build a platform that allows organizers to enter event information once and use it to create, adapt, review, and manage content across multiple platforms.

Expected Deliverable • Single event brief input • Multi-platform content generation • Platform, audience, and tone-based content adaptation • Content editing and regeneration • Event-information consistency checking • Club and brand context • Content quality analysis • Content version management Judging Parameters • Accuracy and consistency of generated content • Quality of platform-specific adaptation • Relevance to the target audience • Effectiveness of consistency checking • Quality of editing and regeneration • Overall usability Technology Parameters AI Models: Gemini, ChatGPT, Claude, Llama, Mistral, etc. APIs: Gemini API, OpenAI API, social/content APIs, other suitable APIs, etc. Frontend: React, Next.js, HTML/CSS/JavaScript, Tailwind CSS, etc. Backend: Node.js, Python, Java, PHP, etc. Database: MongoDB, PostgreSQL, MySQL, SQLite, Firebase, JSON, etc. Other: Document processing, image understanding, analytics, authentication, notifications, etc.

PS 5 — Hobby Matchmaker for Aatmoday. Problem Statement - Students often have interests and hobbies but may not know which Aatmoday sub-group, hobby community, or event would be the right fit for them. This can make it difficult, especially for new students, to discover relevant communities and start conversations. Build a platform where a student simply describes their interests, and the system recommends relevant Aatmoday groups or events. The system should also explain why the recommendation is a good match and generate a personalized icebreaker to help the student start a conversation.

Expected Deliverable • Simple free-form interest input • Aatmoday group and event database • AI-based interest understanding and matching • Relevant group or event recommendations • Short reasoning behind each recommendation • Personalized icebreaker generation • Support for different types of student interests Judging Parameters • Relevance of recommended groups or events • Quality of the match reasoning • Personalization and naturalness of the icebreaker • Performance across different sample interests • Ease of use • Overall functionality of the MVP Technology Parameters AI Models: Gemini, ChatGPT, Claude, Llama, Mistral, etc. APIs: Gemini API, OpenAI API, other suitable APIs, etc. Frontend: React, Next.js, HTML/CSS/JavaScript, Tailwind CSS, etc. Backend: Node.js, Python, Java, PHP, etc. Database: MongoDB, PostgreSQL, MySQL, SQLite, Firebase, JSON, etc. Other: Search, recommendation systems, authentication, notifications, analytics, etc.

PS 6 — Open Innovation. Problem Statement - Not every useful problem can be defined in advance. In this track, teams are free to identify and solve a real problem related to our university or a practical everyday-life problem. The problem can be related to students, faculty, clubs, campus services, administration, or any realistic day-to-day situation. The focus should be on identifying a genuine problem and building a practical technology-based solution for it.

Expected Deliverable • Clearly defined real-world problem • Clearly identified target users • Working software solution or MVP • Complete and usable workflow • Appropriate data handling • Meaningful use of AI/model capabilities wherever relevant • Demonstration of the solution in a realistic use case Judging Parameters • Relevance of the problem • Real-world usefulness • Originality of the solution • Feasibility • Technical implementation • User experience • Quality of AI integration, where applicable • Potential impact and scalability Technology Parameters AI Models: Gemini, ChatGPT, Claude, Llama, Mistral, local models, etc. APIs: Gemini API, OpenAI API, public APIs, third-party APIs, webhooks, SDKs, etc. Frontend: React, Next.js, HTML/CSS/JavaScript, Tailwind CSS, Bootstrap, etc. Backend: Node.js, Python, Java, PHP, FastAPI, Django, etc. Database: MongoDB, PostgreSQL, MySQL, SQLite, Firebase, Supabase, JSON, etc. Other: Authentication, QR, OCR, PDF, maps, notifications, analytics, file processing, etc.