Browse Machine Learning Community (24)

Coryskimming from Microsoft introduces the packed line-up for Azure at KubeCon Europe 2026, spotlighting hands-on AKS labs, AI/ML workload sessions, security, cloud-native DevOps practices, and open-source solutions from Microsoft's top engineers.

damocelj offers a practical walkthrough on securely deploying LLM inferencing with vLLM and NVIDIA NIM microservices in air-gapped Azure Kubernetes Service clusters, tackling network isolation, GPU configuration, and model artifact challenges.

bobmital shares a hands-on playbook for optimizing enterprise LLM inference on Azure, guiding technical teams through architecture, hardware selection, quantization, and model serving best practices across AKS, Ray Serve, and vLLM.

bobmital examines the architectural and economic challenges of large language model inference at enterprise scale, with a focus on Azure and Anyscale’s Ray integration for distributed AI workloads.

bobmital examines the unique challenges of enterprise-scale LLM inference, focusing on the interplay of accuracy, latency, and cost in Azure deployments using Anyscale Ray and AKS. This article provides actionable insights for architects and engineers deploying AI workloads in the cloud.

AnaviNahar introduces Azure Databricks Lakebase, now generally available, highlighting its serverless architecture and AI-native features for building real-time, intelligent applications on Azure.

bobmital presents a comprehensive and practical guide for deploying and optimizing large language model inference on Azure Kubernetes Service, focusing on engineering tradeoffs, GPU efficiency strategies, open-source model evaluation, and robust enterprise security architecture.

Chunlong Yu and co-authors present GenRec Direct Learning (DirL), a Microsoft-driven approach that transforms traditional ranking pipelines by leveraging end-to-end token-native sequence modeling, with experiments and production deployment on Azure Machine Learning.

Yongguang Zhang presents an in-depth view of Microsoft’s AI-powered RAN and intelligent edge strategy, showing how AI, Azure, and advanced platforms are set to revolutionize the future of telecom networks through automation, edge intelligence, and innovative new services.

Sally Dabbah explains how to orchestrate Azure Synapse Analytics pipelines for predictable execution on shared Spark pools. Key techniques include workload prioritization and adaptive orchestration strategies.

AnaviNahar introduces the general availability of Serverless Workspaces in Azure Databricks, detailing their architecture and guidance for when to choose Serverless or Classic models.

NaufalPrawironegoro shares engineering insights on building robust real-time data pipelines with Microsoft Fabric. Learn best practices for data quality, lag management, and operational foundations based on enterprise experiences.

kinfey presents a comprehensive exploration of building an agentic podcast studio using the Microsoft Agent Framework, local SLMs, and VibeVoice. This guide reveals how edge-first AI orchestration empowers privacy, speed, and scalable creative automation.

NaufalPrawironegoro demonstrates an advanced architecture for multi-store retail data ingestion using Microsoft Fabric, Delta Lake, and Azure Event Hubs. The guide explains operational workflows, automation patterns, and best practices for seamless store onboarding.

GeertVanTeylingen presents a comprehensive exploration of the Azure NetApp Files object REST API, demonstrating how it empowers direct analytics and AI access to enterprise file data via S3-compatible object interfaces.

Hannah Abbott details a demo application for healthcare organizations demonstrating how Azure AI powers real-time, secure transcription and clinical text analytics. Developed alongside Samuel Tauil, this solution helps teams streamline data workflows and analytics using Microsoft cloud technology.

Pamela Fox invites you to a free, six-part livestream series on building advanced AI agents in Python using the Microsoft Agent Framework. The series covers agent fundamentals, memory, RAG, workflow orchestration, monitoring, and HITL workflows with live demos and practical examples.

JohnGruszczyk discusses how Insilico Medicine’s Nach01 model integrates with Microsoft Discovery on Azure, enabling AI-driven, reproducible, and scalable drug discovery workflows for researchers.

lmiroslaw showcases how Neural Concept utilized Azure HPC and AI infrastructure to achieve record-setting accuracy and efficiency for industrial aerodynamic workflows, leveraging massive datasets and advanced machine learning techniques for real-world automotive impact.

Rafia Aqil and co-authors present a comprehensive technical guide to disaster recovery for Azure Databricks and Microsoft Fabric, focusing on automation, cloud resiliency, and best practices for DR in analytics platforms.