Generative AI made Easy with Google Cloud
I'd recently read a post stating that "using Google AI is difficult" at least more difficult than the competition. I have to admit I found this mildly amusing. This is one of the reasons I've been publishing additional examples on this subject. I recently had a meeting with one of the largest grocery chains here in the US, the opportunity further underscored this common misbelief. Thankfully the leadership I was meeting with had an amazing technical acumen and we walked through several example and I quote: "But wait, where is the complexity?" at which I replied: "It's only complicated if you choose to make it complicated". Further explaining that we do add complexity when integrating with legacy systems, but adding new functionality DOES NOT need to incur high technical debt.
Simply put, if you want to use an intelligent Generative AI client, you can do so in JavaScript, Python, Java, and Go by simply using the provided Google Libraries. If the Library is too much to digest, your development team can call the API directly using any HTTP client.
Examples:
Complete E2E Catalog Enrichment:
This example contains an Express middle tier and React front-end demonstrating: Zero Shot, Multi-Shot, Chain of Though, Batch / Stream Processing, Voice and Video integration, using Google as a grounding model, Imagen, and much more:
Python Simple Agent and Local Retrieval Augmented Examples: