Cloud and DevOps engineers. Azure is the second-largest cloud in the world, and engineers in Hyderabad and Bengaluru build and run parts of it. Cloud and DevOps engineers here work on cloud-native platforms, on site reliability, the work of keeping a service running and fixing it fast when it fails, and on the servers and networks underneath. These are programming jobs as much as operations jobs. Most cloud postings require Java and Python, Java about twice as often as cloud postings at product companies, many require JavaScript or C/C++, and some require Go, alongside Azure, Kubernetes, Linux and PowerShell. Many also require knowledge of security and of the infrastructure behind AI. This is Microsoft's largest role, about twice its weight across product companies, and an unusually senior one, with more staff-level openings than any other role here.

GenAI engineers. Microsoft is putting Copilot, its AI assistant, into Windows, Microsoft 365, Azure, Dynamics and its developer tools, and GenAI engineers build those features. Most of the work is fitting language models into existing products rather than building new AI products from scratch. The postings require Python above all, with Java, JavaScript and TypeScript for the products around the model, and React for its screens. Some require Azure OpenAI and Azure AI Foundry, Microsoft's services for using and building on language models, with semantic and vector search for finding the right information to give a model. A notable share require Responsible AI, the practice of checking that AI features are safe, fair and explainable, which appears in Microsoft's GenAI postings more than twice as often as in GenAI postings at product companies. GenAI engineers are about twice as common here as at a typical product company, and their share of openings has grown in each quarter since January to March 2026.

Data engineers. Microsoft sells data platforms, Microsoft Fabric, Azure Synapse, Azure Data Factory and Power BI, and it runs some of the largest data pipelines in the world inside its own products. Data engineers here build both. Many postings require Microsoft Fabric itself, with Spark, Kafka and streaming, and most require Python, often with Java or C/C++. Data engineers are about twice their usual weight here.

C/C++ systems engineers. Beneath Azure sits a layer of software close to the hardware, which runs virtual machines, moves packets across networks and stores data on disk. C/C++ systems engineers on Linux build it. The postings require C/C++ with networking, TCP/IP, DHCP and DNS, file systems and device drivers, and a few require Hyper-V, Microsoft's software for running many virtual machines on one server. Some require Rust, which Microsoft is adopting for new systems code because it prevents many kinds of memory errors. This role is well above its usual weight.

.NET engineers. .NET is Microsoft's own platform, so it is perhaps surprising that .NET engineers appear here at only about their usual weight. One likely reason is that C# appears in most Microsoft postings, whatever the role, so much of Microsoft's .NET work sits under other roles. The postings for this role require C# with Azure, React, Angular and TypeScript, and many also require Java and Python. A good share require experience of building with language models. .NET's share of openings rose during 2026.

Software Engineer, Security. Software engineers who specialise in security: they build login systems, cloud and network protections, and the checks that keep code safe. Microsoft sells security products, Defender, Sentinel and Entra among them, and has to protect its own cloud. Security software engineers here are about twice their usual weight, and most are application security engineers, who find and fix weaknesses in code. The postings require C/C++ and Java with cryptography, and many require Prisma Cloud, a tool for checking that cloud systems are set up safely, and some require OpenTelemetry for tracing what a service does. This is among the most senior roles at Microsoft.

Java backend and full-stack engineers. Java engineers build services behind Microsoft's products, with Spring Cloud, Azure Service Bus and Azure. Nearly every Java posting at Microsoft also requires C#, most also require Python and JavaScript, and many require experience of building with language models. Java is below its usual weight here, and its share of openings has fallen steeply through 2026, to very little by July to September.

Power Platform and business application developers. Microsoft builds the Power Platform, Power Apps for building simple business apps and Power Automate for automating tasks, and it sells Dynamics, its software for a company's finance, operations and sales. Developers for the Power Platform appear at several times their weight across product companies. The postings require C#, Java and JavaScript with Azure, Dataverse, the Power Platform's own database, and Power Apps and Power Automate. Developers for the business applications, at about their usual weight, require C#, Java and JavaScript with Azure and Power BI. Power Platform jobs fill unusually quickly.

Embedded firmware engineers. Microsoft designs hardware, from the boards and chips inside Azure's servers to its Surface computers and Xbox consoles, and that hardware needs firmware, the software that runs directly on a chip. Embedded engineers write it in C/C++ with device drivers and the standard connections between chips and devices, I2C, SPI, PCIe and USB. Some postings require Secure Boot, which checks that a machine starts only trusted software, and some require ONNX or TFLite for running AI models on a device. Embedded engineers are a little below their usual weight, but their share of openings grew through 2026.

Generalist software engineers. A group of postings describe general development work rather than one stack. Most of them require C#, Java, Python and JavaScript together, often with C/C++, and a few require WPF, Microsoft's framework for Windows desktop applications. Others describe backend work in any language. Together they are above their usual weight.

ML engineers. A small group of ML engineers train and adapt models, mostly for language. The postings require Python with PyTorch and Transformers, and a few require CUDA, NVIDIA's platform for running models on graphics chips. They are above their usual weight and among the roles most open to juniors here.

Also present. Frontend engineers build Microsoft's web screens in React and TypeScript, with accessibility, making screens usable by people with disabilities, a recurring requirement. They are well below their usual weight. Mobile engineers work on Android and iOS, a little below their usual weight. Full-stack engineers and Go backend engineers appear in small numbers, well below their usual weight. A few SAP engineers work on Microsoft's own SAP systems, some integration developers connect them to other systems, and a handful of chip designers work in FPGA and SystemVerilog. Test engineers are a small fraction of their usual weight.

Rare here. Python backend engineers, JavaScript and Node full-stack engineers, and Salesforce and ServiceNow engineers barely appear.