Aman Jain on LinkedIn: Jayu | Gemini API Developer Competition | Google AI for Developers (2024)

🚀 Test Our Fun New Anonymous Messaging Platform! 🚀Link- : https://lnkd.in/d6UhFAgEWe're excited to introduce our latest project: an Anonymous Messaging Platform built with Next.js and powered by Google Gemini AI. This platform lets you send anonymous messages to anyone with a unique link, making it a fun and engaging way to communicate.🔹 Key Features:Anonymous Messaging: Send and receive messages anonymously through unique links.AI Insights: Enjoy fun AI-generated insights and responses.User-Friendly Design: Experience a sleek and intuitive interface for seamless interaction.Robust Security: Your anonymity and data privacy are fully protected.🔹 Why Try It? This platform is perfect for fun, casual interactions, and anonymous feedback. Share your thoughts without revealing your identity and explore AI-driven responses.🔹 Development Highlights:Framework: Built with Next.js for dynamic server-side rendering and static site generation.API Integration: Seamlessly integrated with backend services for efficient data handling.Responsive Design: Optimized for both desktop and mobile devices.Join our testing phase and share your anonymous messages today! Your feedback will help us enhance the platform. 🌟Try it out now!#AnonymousMessaging #AIInsights #Nextjs #GoogleGeminiAI #FunCommunication #TechInnovation #FullStackDevelopment #UserExperience #Security #BetaTesting #UniqueLinks

Just tried the new Gemini 1.5 Flashmodel! Gemini is designed to solve complex reasoning problems with an impressive balance of flexibility, speed, and cost efficiency. Whether you’re in tech, finance, or any other industry, the Gemini 1.5 Flash model promises to revolutionize how we approach problem-solving and innovation. It’s a game-changer for anyone looking to leverage AI for efficient, faster, and more cost-effective solutions.Here is my attempt with Javascript - React JSLive demo: https://lnkd.in/gFRaK8MHAPI integration: https://ai.google/build#Gemini15Flash

11

1 Comment

Here is an open-source library that makes integrating AI into an application extremely easy.CopilotKit. Star their repository: https://lnkd.in/eHxrtNVECopilotKit will take your application context and feed it into their React infrastructure to build:• In-app AI chatbots• AI-powered textareas• RAG, function calling, and integrationsThe library has native support for LangChain, LangGraph, and LangServe. You can use this to extend the engine's skills.It also provides built-in native UI/UX components you can use as part of your applications:• CopilotChat• CopilotSidebar• CopilotPopup• CopilotTextareaThe library is open-source. You can self-host it. You can use it with any LLM, including GPT-4.This project was #2 on HackerNews and ProductHunt. It was trending in GitHub.This library works on any React app, but the team is working to expand it.Thanks to the team for showing me their tool and collaborating with me on this post!

195

9 Comments

✨✨✨On Google I/O Keynotes several hours ago, certainly I was engrossed throughout, in the list including:• Gemini Pro• Gemini Flash• Gemma• PaliGemma• NotebookLM• Project Astra• Veo• Imagen 3The main keynotes being on AI, they are also befitting of Workspace, Vertex AI and AI Studio; the 3 that I commit.In tandem, I ended wanting to witness more on Gemini AI relevance specific to Google’s existing frontrunner technologies not mentioned in this year’s Google I/O.Generally, I’m piqued in applications on: • Gemini AI in Google Home and Google Nest given that they are top-tier on personal smart devices; • Gemini AI relevant on Youtube given Youtube is by and large the most advanced search, create and streaming platform expanded to YoutubeVR and Youtube360;• Gemini AI for Google Maps and StreetView and Earth given their respective gold-standard;• AI for Flutter given that progressive web apps dev go faster and more streamlined.• GoLang and Kotlin.• Gemini AI for Google Workspace for Education, given that Practice Sets is adaptive learning.• How about Gemini AI on reviving my favorites but long discontinued Google Poly and Google Tiltbrush?Regardless the above, Google I/O 2024 covered a diverse gamut of AI advancements. Considering that I do Go, Android, Swift and Flutter, and did end up joint-85th on the second last Google CodeJam, it’s for me to drive “my” DeLorean, with Astra specs worn.#gemini #gemma #paligemma #projectastra #veo #imagen3 #notebookLM #AI #GenAI #responsibleAI

5

🚀 Exciting News for Developers: Google Unveils AI-Powered Debugging in Chrome DevTools! 🛠️💡Are you tired of deciphering cryptic error messages and debugging web apps? Good news! Google is revolutionizing the way developers tackle these challenges with a new Gemini-powered feature in Chrome DevTools.Starting this week, you'll experience a smarter way to debug with personalized, contextual AI assistance right in the Chrome DevTools Console. Say goodbye to the frustration of interpreting errors and hello to clear explanations and actionable solutions, courtesy of cutting-edge AI technology.👨💻 John Dahlke, Google's Product Management Director, shared insights at a recent press briefing: "Chrome DevTools is already a go-to for developers to refine their apps. Now, we're enhancing it with AI to make error messages more understandable and provide developers with suggested fixes."This innovation is not just about convenience; it's about empowering you to focus on what you do best—creating and innovating. Let Gemini handle the troubleshooting so you can reclaim your time and energy for more important tasks.🤖💼 For teams dealing with unfamiliar code, this AI feature is a game-changer. Quickly get up to speed with Gemini's insights, pinpoint issues, and move forward with confidence.Stay tuned for more updates, and get ready to transform your debugging experience with Google's AI-powered Chrome DevTools!#Google #ChromeDevTools #AI #Debugging #WebDevelopment #Innovation #DeveloperTools #GeminiAI

3

Let's talk about AI, VR and captions!As my hearing has declined, captions have become increasingly important to me. I rely on Google's captioning technology on a daily basis for meetings, phone calls, and even scrolling through Instagram. Google's Live Transcribe app uses an ML model that runs locally on my phone to transcribe audio lightning fast, with pretty decent results. However, there are some limitations: unfamiliar words can trip it up (tldraw often becomes "children," for example), and there's no speaker recognition, which means when there are overlapping voices, you have to scan through a stream of text. Additionally, looking down at your phone can feel antisocial, while holding it up by the speaker's face is awkward. 😬How can we improve on this situation? OpenAi's Whisper can be taught unfamiliar words, I have a project on the go at the moment to use the SpeechMatics API (a kind of websocket wrapper around Whisper) to do just that. However, running the model locally would be ideal, so with iteration 2 I'm going to look at ratchet, an ML browser runtime developed by a fellow Londoner which currently supports whisper https://lnkd.in/e75eUTGK Speaker recognition is a problem that would have to be solved in the training of the model itself. Intriguingly, the source code for Live Transcribe makes reference to speaker recognition, so it may be a feature they roll out soon. https://lnkd.in/ex78Gsu8As for holding the phone, I've come to the conclusion that Google Glass was probably too ahead of its time! It would be amazing to have a lightweight overlay on the world with captions streaming at the bottom of the screen. Iteration 3 of my transcription app might involve hacking away at one of those.It feels like all the prerequisites for the captioning app of our dreams have been fulfilled, but it's not here yet. I guess accessibility apps just don't have the same allure as AI girlfriends or automated workers.

10

🌟Exciting News!🌟I’m thrilled to share that I’ve earned the Leveraging the Gemini Pro Vision model for image understanding, multimodal prompts, and accessibility prestigiousGemini Pro Vision Badgefrom Google for Developers! 🚀🔍📸Thrilled to share that I've completed the training on Google's Gemini Pro Vision model, delving into the cutting-edge realms of image understanding, multimodal prompts, and accessibility. 🌐💡🔍What’s the Gemini Pro Vision model?The Gemini Pro Vision model is a cutting-edge AI model developed by Google. It’s designed to handlemultimodal input data, including bothtext and image prompts. With this powerful model, we can unlock new possibilities in image understanding, accessibility, and more! 🌐📷🌐How I Leveraged Gemini Pro Vision:I dived deep into the world of Gemini, exploring its capabilities and pushing boundaries. From analyzing HTML documents to enhancing image descriptions, I harnessed the power of Gemini to make the digital world more accessible. 🌈🌐🌟Why It Matters:As developers, we play a crucial role in creating inclusive experiences. By leveraging Gemini Pro Vision, we can enhance web accessibility, improve multimodal interactions, and create a more connected digital ecosystem. Let’s build a future where everyone can participate! 🌎🤝🚀Next Steps:I’m excited to continue my journey with Gemini. Stay tuned for more updates, tutorials, and real-world applications! 📚👩💻#GoogleDevelopers #AI #GeminiProVision #Accessibility #MultimodalMagic

9

🚀 Excited to announce the launch of Visu-Ai: The Future of AI Video Generation! 🎥✨I've just released my latest project, Visu-Ai, a cutting-edge app that harnesses the power of AI to revolutionize video creation. 🛠️ Tech Stack:• Next.js for a smooth, responsive frontend• Gemini API for advanced AI processing• Google Text-to-Speech for natural voiceovers• Assembly AI for precise caption generation• Replicate for stunning image creation• Remotion for seamless video composition• PostgreSQL & NeonDB for robust data management🌟 Features:• AI-powered video generation• High-quality text-to-speech conversion• Automatic captioning for accessibility• Custom image creation integrated into videos👀 See it in action: https://lnkd.in/gHctXS3J🔍 Curious about the code? Check out the GitHub repo: https://lnkd.in/g-EYn9tnThis project represents countless hours of coding, testing, and refining. I'm thrilled to share it with the LinkedIn community and can't wait to see how it transforms video creation for creators, marketers, and businesses alike.Have you experimented with AI in video production? I'd love to hear your thoughts and experiences in the comments below!#AIVideoGeneration #TechInnovation #MachineLearning #VideoProduction #NextJS #ArtificialIntelligence

24

2 Comments

🚀 Excited to Share My Latest Project! 🚀I am thrilled to announce that I have developed a new project, the Gemini AI Health App, and it's now live at https://lnkd.in/dWueEXSfProject Highlights:Technologies Used:Python: The core language for development.Streamlit: For creating a seamless and interactive web app experience.Google Gemini Pro Vision API: Leveraging state-of-the-art AI for generative content.Pillow: For handling image processing tasks.dotenv: To manage environment variables securely.Key Features:AI-Powered Health Insights: Utilizing Google's advanced AI capabilities to provide meaningful health insights.Interactive Interface: Streamlit ensures the app is user-friendly and interactive.Secure and Scalable: Robust integration of environment variables for secure API key management.Achievements:Successfully implemented AI models to enhance user experience and provide accurate health insights.Created an intuitive interface that simplifies complex health data for end-users.I am proud of the progress and the potential impact this app can have. Looking forward to your feedback and suggestions!Check out the app here: https://lnkd.in/dWueEXSf#AI #HealthTech #DataScience #Streamlit #Python #GoogleAPI #MachineLearning

14

1 Comment