Principal AI Software Engineer

Overview

We are building the future of security posture and exposure management at Microsoft, protecting millions of enterprise customers across cloud, device, identity, data, and AI workloads. Our team is shifting from legacy dashboard experiences to AI-native, agentic security that drives risk reduction at scale.

We are looking for a Principal Software Engineer at the intersection of AI/ML and security. You will drive the architecture and delivery of autonomous agentic experiences that investigate, triage, remediate, and harden customer environments using LLMs, knowledge graphs, and the Microsoft Security Graph (MSG).

This is a high-impact IC role reporting to the Exposure Management engineering leadership. You will influence strategy, mentor senior engineers, and ship production systems used by the largest enterprises in the world.

Security is entering its agentic era. Legacy dashboards and manual triage cannot keep pace with AI-powered threats. Microsoft protects more enterprise surface area than any other vendor, 65M+ devices, 3 clouds, 750M+ identities, and 78T daily signals.

You will architect AI systems that turn this data into autonomous action, agents that find what matters, explain why it matters, and fix it in minutes, not months.

Responsibilities

Architect & Build Agentic Security Experiences

Design and implement Blue (investigate/triage), Green (remediate/harden), and Red (attack simulation) agents operating across customer environments
Define agent architecture: planning, tool use (MCP skills), memory, context retrieval, and human-in-the-loop controls
Build orchestration over the Microsoft Security Graph, correlating exposure signals across cloud, device, identity, data, and AI

Drive AI-Native Platform Design

Architect skill, knowledge, context, and memory layers for agentic security systems
Design MCP-based skill interfaces for customization and extensibility
Define how LLMs interact safely and accurately with structured security data (vulnerabilities, misconfigurations, attack paths, graph relationships)

Shape AI-for-Security Strategy

Partner with PMs and domain leads to define where AI autonomy vs. human involvement is needed
Evaluate and integrate foundation models (Azure OpenAI, fine-tuned models) for tasks such as risk scoring, remediation planning, and blast radius analysis
Stay ahead of industry solutions (Wiz AI-APP, CrowdStrike Charlotte AI, Palo Alto XSIAM)

Technical Leadership & Influence

Set technical direction across multiple teams (20–30 engineers)
Drive architecture decisions, design reviews, and engineering excellence
Mentor senior engineers (L63–L65) and grow the AI-for-security discipline
Represent the team across Microsoft (Security Copilot, MSG, Azure AI, Research)

Qualifications

Must Have

8+ years of software engineering experience, with 3+ years applying ML/AI to production systems
Deep expertise in LLM application development, including prompt engineering, RAG, and agent frameworks such as AutoGen, Semantic Kernel, LangChain, or similar
Strong systems design skills, including distributed cloud systems, streaming pipelines, graph databases, and API design at scale
Experience building security products or working with security data such as vulnerabilities, misconfigurations, identity, and cloud posture
Track record of driving ambiguous, cross-team technical initiatives from concept to production

Preferred

Proficiency in Python and at least one systems language such as C, Go, Rust, or Java
Experience with Microsoft Security stack including Defender, Sentinel, Entra, Purview, Intune, and MSEM
Familiarity with MCP (Model Context Protocol) and tool-use patterns for LLM agents
Background in security operations, red teaming, or exposure management
Experience with knowledge graphs and graph-based reasoning for security
Publications or patents in AI/ML applied to cybersecurity
Experience with Azure OpenAI Service, Copilot extensibility, or Security Copilot skills development
BS or MS in Computer Science, Engineering, or equivalent experience

This position will be open for a minimum of 5 days, with applications accepted on an ongoing basis until the position is filled.

Microsoft is an equal opportunity employer. All qualified applicants will receive consideration for employment without regard to age, ancestry, citizenship, color, family or medical care leave, gender identity or expression, genetic information, immigration status, marital status, medical condition, national origin, physical or mental disability, political affiliation, protected veteran or military status, race, ethnicity, religion, sex (including pregnancy), sexual orientation, or any other characteristic protected by applicable local laws, regulations and ordinances. If you need assistance with religious accommodations and/or a reasonable accommodation due to a disability during the application process.