Lauren Smith

Head of Growth @origon
Sharing AI, tech & startup learnings as they happen.

[Lauren Smith @LearningwLolo]

I'm reading through Google's Gemini Embedding 2 release and here's what I think it actually opens up for anyone building AI systems.

Embedding models are what let AI systems search through data by meaning instead of keywords.

They turn your text/images/videos into numbers (vectors) that AI can compare.

Before this, if you were building an AI system (like a chatbot that answers questions using your company's documents, videos, and images) you needed separate embedding models for each type of data.

You'd embed all your text with one model. Images with a different one. Videos with another. Audio with yet another.

Four different embedding spaces. Four different systems to maintain.

Gemini Embedding 2 collapses all of that into one model.

Which (if I'm understanding this correctly) means if you're building an AI assistant and someone asks about "labradoodle," the system can now pull from:
- Text documents mentioning labradoodles
- Photos of labradoodles
- Videos of them playing
- Audio of them barking
All from one unified embedding space.

My dog is sitting behind me as I'm writing this (hence the labradoodle reference). When I think about him, I don't separate "visual memory" from "audio memory" from "text description." I just think about him, all of it at once.

Gemini Embedding 2 treats text, images, video, and audio as different expressions of the same underlying meaning.

Which is how humans have always thought.

We just accepted for years that AI systems couldn't work that way. That you had to build separate infrastructure for each modality. Separate pipelines. Separate teams.

I'm reading this thinking, we don't have to do that anymore? 😅

I don't know where this goes, but it feels like we just removed a pretty fundamental limitation in how we build AI systems.

[Google AI Developers @googleaidevs]

Start building with Gemini Embedding 2, our most capable and first fully multimodal embedding model built on the Gemini architecture. Now available in preview via the Gemini API and in Vertex AI.

[Lauren Smith @LearningwLolo]

@ATC_SECURE @bindureddy what have you switched to? i still find opus and sonnet to be fine as long as i'm creating new chats to keep the context window manageable

[Lauren Smith @LearningwLolo]

@krishnarajr_01 @origon Hi Krishna, just trying to understand that you have built a flow before engaging. Not trying to steal ideas

[Lauren Smith @LearningwLolo]

Calling all AI automation builders

If you've built workflows on n8n, LangChain, or Make then we want to work with you!

Create templates on @origon, get paid, and get published in our marketplace with full credit + promotion

Interested? Repost + comment what you've built 👇🏼

[Lauren Smith @LearningwLolo]

@_harjaapdhillon @origon What platform/framework did you build it on top of??

[Lauren Smith @LearningwLolo]

@codewithimanshu @origon I know! I would love to replicate some of that for Origon

[Lauren Smith @LearningwLolo]

@iusemedia @origon I'd love to see it!

[Lauren Smith @LearningwLolo]

We're building a developer community at Origon for people who love creating useful systems.

If that's you, drop what you've built below. Could be n8n workflows, LangChain apps, Make automations, anything in this space.

Let's build together!

[Lauren Smith @LearningwLolo]

what we're looking for: developers who've built real automations that companies use

the stuff that's actually running in production solving problems

[Lauren Smith @LearningwLolo]

@FirstFloris @origon Messaged you

[Lauren Smith @LearningwLolo]

I’m looking for a Motion Designer to help us scale @origon 🚀

We need someone to turn our Figma designs into promo and animations for tutorial content.

Think ElevenLabs, Linear, or Framer aesthetic.

Interested? Drop your portfolio/showreel below! 👇

[Lauren Smith @LearningwLolo]

@Dash_FX @origon Have you done full videos as well? (45-60 sec)

[Lauren Smith @LearningwLolo]

@evelyn_artt @origon Mostly for short form promos, but we'll also need an editor for full tutorials as well

[Lauren Smith @LearningwLolo]

@zeeshan_zag @origon doesn't look like the 'schedule meeting' works on your website right now

[Lauren Smith @LearningwLolo]

@byshubh_ @origon The link in your bio doesn't go anywhere. Do you have a portfolio or site?

[Lauren Smith @LearningwLolo]

@Wic_William Not checking DMs, please reply here for me to see it. Thanks!

[Lauren Smith @LearningwLolo]

@w4jhiflx @origon I'm not checking DMs, please post your portfolio here for me to see it. Thanks!

[Lauren Smith @LearningwLolo]

@suayrez @origon Do you have a portfolio that I could look at?

[Lauren Smith @LearningwLolo]

@J_wiil7 @origon Have you done full animated videos before? I didn't see any on your behance

[Lauren Smith @LearningwLolo]

@Alex_designerr @origon We're looking to do several different projects. Feature videoa, explainers, etc.