Back from a week in Berlin full of conferences. Had a great time meeting new people, MCing some of my industry idols and attending great talks at @BerlinBuzzwords & a blast organising the second edition of @BeamSummit with this great team. Hope to see you all again soon!

I've written something! About AI and LLMs 🤖! And how to make 'em fast 🏃 & save cost 💰. Sneak peak: @GroqInc helped a lot. Short post here: Long post here: Let me know what you think!.

Hey @madebygoogle @googledrive - the back-up of my old phone is gone and Support can't seem to help me. Looking for an explanation of why my backup would be deleted and how we can avoid it in the future. Thanks!.

RT @ApacheBeam: Join us on the 7th edition of this awaited event for professionals in real-time processing, data, and AI/ML engineering 🌟 #….

RT @johnrhanger: Good morning with good news: "China's clean power expansion is now pushing coal into reverse." . China added 102 GW of sol….

RT @ollama: Ollama 0.2 is here! Concurrency is now enabled by default. This unlocks 2 major features: . Parallel….

RT @lulumeservey: Yesterday I saw the wildest CEO talk maybe ever, and finally learned what Tobi was building that Saturday in May. So Sho….

RT @BeamSummit: If you want to become a speaker at #BeamSummit 2024, you are still on time to submit your proposal! Don't miss the chance t….

RT @paulg: I don't think I'm ever calmer than when scuba diving.

Ok this is pretty cool.

RT @waitbutwhy: Five cities, then and now. Somewhere today there's a quiet fishing village that will be a 2100 megalopolis. Hong Kong: 196….

I was especially impressed by the GenAI @lavinigam and @iamthuya in which they presented how you can feed any document (think insurance contract or employment contract) and ask it natural language questions. Exciting times!.

Personal take-aways of things I need to try after the conference include:.- Throw each ChatGPT query at Bard as well and compare. - Use Google’s APIs that give you access to their LLM models. - All the goodies that come with Cloud Code; automatic code completion and documentation.

I am keen to see how Google will (further) integrate the Bard results with its existing knowledge graph - but am also aware I’ll need a lot more experimenting across tools to have a more well-rounded opinion.

Particular things to note in the two answers below would be the “freshness” of the data and the out of the box integrations Google offers towards its product portfolio (look at that beautiful table format!).

A lot of chatter & chirping about LLMs & Generative AI (as expected) at #GoogleIOConnect. I have been dabbling & enjoying my time with #ChatGPT quite a bit, but have a feeling we can’t write off Google’s entries in the same space (and spread over its whole portfolio of products).

RT @ApacheBeam: 👋🎉 Thank you for being a part of the amazing Beam Summit 2023! .🌟 We hope you had a blast exploring the world of data and g….

RT @ApacheBeam: Loving the view from #BeamSummit

RT @alexvb: Next week is @BeamSummit. Join the community conversation on the "Future of @ApacheBeam". discuss missing features or what th….

RT @ApacheBeam: 🗣️ Take a trip down memory lane with Kenn Knowles, Robert Bradshaw, and Reuven Lax as they reflect on their early experienc….