engblogs

FabCon 2026 showcases Azure Databricks' Lakeflow, Lakebase, and Genie innovations—unifying data ingestion, operational data, and AI-assisted analytics on an open lakehouse foundation with deep Microsoft 365 integrations.

Part 1 shows how to build a modular Multi-Agent System (Dev Signal) that discovers Reddit questions, grounds findings with Google Cloud docs via MCP and the Google ADK, and generates blogs and visuals, deployed to Google Cloud Run with a long-term memory layer for future expansion.

Practical guide to scaling read workloads with Cloud SQL read pools and autoscaling, enabling a single read endpoint, automatic node adjustments, and cost-efficient high availability.

GA of Memorystore for Valkey 9.0 delivers massive throughput and lower latency with SIMD optimizations, pipeline memory prefetching, new HEXPIRE commands, geospatial querying, and cluster-enabled multi-database support for scalable, managed caching.

Orchestrates a distributed team of specialized AI agents—researcher, judge, and orchestrator—into scalable microservices deployed on Cloud Run, enabling seamless frontend integration via the ADK and A2A protocol.

Nova Forge SDK unifies end-to-end LLM customization for enterprises, enabling developers to prepare data, train models, and deploy Nova variants across Amazon Bedrock and Amazon SageMaker AI with guided workflows and scalable automation.

An end-to-end, Nova Forge SDK–driven guide to LLM customization—from data loading and supervised fine-tuning (SFT) to reinforcement fine-tuning (RFT), evaluation on a Stack Overflow classifier, and deployment on SageMaker AI Inference.

An end-to-end, LLM-driven pipeline that translates natural-language access policies into executable Rego (OPA policy-as-code), featuring policy detection, extraction, schema validation, linting, compilation and automated testing, demonstrated on the ACRE dataset with high compile and test pass rates to enable Zero Trust and auditable policy enforcement.

End-to-end guidance for migrating from Amazon Nova 1 to Nova 2 Lite on Amazon Bedrock, detailing feature upgrades, API changes, migration steps, and production-ready evaluation to maximize reasoning, tool use, and throughput.

How Bark.com and AWS engineered a scalable, AI-powered video generation pipeline using SageMaker, Bedrock, and multi-GPU inference to automate creative ideation, ensure brand consistency, and accelerate content production.

AI-powered, adaptive A/B testing on AWS using Amazon Bedrock and MCP to deliver real-time personalization with a serverless, scalable architecture.

A practical guide to production-ready AI agent evaluation with Strands Evals, showing how to define cases and experiments, deploy LLM-based evaluators, and validate online and offline, multi-turn agent behavior.