André Bandarra’s Post

👩💻Google company has announced many new AI products and improvements: – Gemini 1.5 Pro now has a context window of 2 million tokens. This is 16 times more than GPT-4o📊 – New AI options for the Google search engine: quick answer to a user’s question, video search, circle and find function (can solve mathematical problems)🎓 Join the world of AI - we will help you take your first steps!

People give Google shit for being behind in AI. I just started using Gemini built into Google Docs and it is really impressive. It's going to fundamentally change how we all use docs in our work. It can do things like summarize etc etc. But you can ask it questions without links to anything and it will search back over all docs, find relevant things, extract out valuable information, add it to its synthesis and so on. This is just the beginning.

In a notable shift within the tech landscape, Google has experienced the departure of another influential figure in the AI domain, as the creator of Keras bids farewell. This move raises important questions about the future of AI development at Google and the broader implications for the industry. For an in-depth analysis of this transition and its potential impact, I invite you to read the full blog post here: [Google loses yet another AI pioneer as Keras creator leaves](https://ift.tt/w5qYRg1). Stay informed about the evolving dynamics in AI and its significant players.

In this week's article we discussed the newest addition to the AI race, Google's AI Overviews. Despite having gotten off to a rocky start, this could be a game-changer for the search engine domain and help Google withstand the intense competition from AI companies. If this sounds interesting, we promise the complete article is even better! You can check it out on the link given below- https://lnkd.in/gRdepuDs For more such updates delivered right into your inbox, comment down your email IDs and we'll add you to our subscriber list :)

Unlock the Power of AI with Google's AI Studio: A Revolutionary Tool for Prompt Engineering. In the ever-evolving world of artificial intelligence, mastering prompt engineering has become a crucial skill. With the introduction of Google's AI Studio, users now have access to an innovative platform designed to enhance their AI capabilities, offering a streamlined interface for crafting effective prompts and generating superior results. What is Google's AI Studio? Discover te response through this article : https://lnkd.in/ee6-g4PD

Unlock the Power of AI with Google's AI Studio: A Revolutionary Tool for Prompt Engineering. In the ever-evolving world of artificial intelligence, mastering prompt engineering has become a crucial skill. With the introduction of Google's AI Studio, users now have access to an innovative platform designed to enhance their AI capabilities, offering a streamlined interface for crafting effective prompts and generating superior results. What is Google's AI Studio? Discover te response through this article : https://lnkd.in/ee6-g4PD

Unlock the Power of AI with Google's AI Studio: A Revolutionary Tool for Prompt Engineering. In the ever-evolving world of artificial intelligence, mastering prompt engineering has become a crucial skill. With the introduction of Google's AI Studio, users now have access to an innovative platform designed to enhance their AI capabilities, offering a streamlined interface for crafting effective prompts and generating superior results. What is Google's AI Studio? Discover te response through this article : https://lnkd.in/ee6-g4PD

Nice work on an #AI #voice assistant with persistent memory from Jakub Neander using Deepgram and #Llama3. In parallel to improving latency, I would introduce some filler words or even phrases during silence, like "let me think about this" or "oh, good question" (in case it is a question, which could be recognised by the question mark of the stt). I mean, is it really a problem that I have to wait a few seconds for a response from the bot? I think the problem is more the silence that irritates me because I can't tell if the bot has crashed or if it's continuing. So let's think more broadly about problem solving, rather than getting too deep into the technology.

💡 Playing with ideas for an AI voice assistant with persistent memory My first attempt, built with Deepgram Groq and Llama3 from AI at Meta It has memory so it has the discussion context from previous exchanges I still need to improve the latency, but it's pretty good already! Interruptions could be more fluid though What features would you like to see?

Big News from Google AI: Gemini 1.5 is Here! 🎉 Just yesterday, Google announced the next generation of their powerful language models: Gemini 1.5. With enhanced capabilities and a groundbreaking 1 million token context window, these models are set to revolutionize how we interact with AI. So, how do you harness this incredible power? It all starts with the prompt! Think of it like giving directions to a super-smart but slightly clueless friend (we all have one 🤪). Here's the secret sauce for getting the most out of Gemini (and any LLM): Be Specific: "Write a haiku about pizza" beats "Write a poem" any day. 🍕 Set the Vibe: Want something serious, silly, or somewhere in between? Let Gemini know! 🎭 Give Examples: Show Gemini what you like! Think of it as giving a mood board. 🖼️ Nail the prompt, and you'll unlock the full potential of AI ✨