A few days ago, I received an invitation from Google Cloud to attend their AI Labs – Ahmedabad event — and this was nothing like a typical tech meetup.

This wasn’t a “sit-and-listen” conference. We were inside the lab, building, testing, and breaking things hands-on.

What made it even more powerful was the diverse group of high-level minds from across the globe — engineers, researchers, and founders working on GenAI at scale.
The knowledge exchange and networking inside that room felt like an acceleration chamber for GenAI innovation.

Labs & frameworks we worked on:

• SLM vs LLM stress testing — when smaller models actually outperform the giants

• Agent-to-agent communication via ADK & MCP protocols

• Prompt routing, grounding, and tool-calling inside Vertex AI

• Agent evaluation using structured scoring & feedback loops

• Orchestrating full workflows using MCP Toolbox & Model Context Protocol

• Understanding Google Cloud’s embeddings, pipelines & data infra in practice

Having previously collaborated with Google Cloud / Google Labs, this deep dive helped connect the dots between AI theory and scalable production design. It wasn’t just learning — it was engineering clarity.

Massive thanks to Google Cloud x Hack2skill for bringing together such a global brain network, and to Aditya Ghanekar and Romin Irani for turning complex systems into executable strategy.

#GenAIExchange #GoogleCloud #AI #VertexAI #LLMOps #AIEngineering