Principal Software Engineering Manager | Ads Trust & Safety AI Platform

Microsoft Advertising serves billions of ad decisions every day across search, native, shopping, display, audience, and AI-powered advertising experiences. Trust & Safety is foundational to this marketplace: protecting users from harmful, deceptive, and low-quality content; protecting advertisers, publishers, and partners from fraud and abuse; and ensuring Microsoft Ads operates with strong policy, privacy, security, compliance, and regulatory integrity.

We are seeking a Principal Software Engineering Manager to lead a key engineering areas within the Ads Trust & Safety AI Platform. This team builds platform capabilities for advertiser and domain intelligence, AI-assisted investigation, model serving, decisioning, enforcement, human review, observability, and risk/fraud protection.

This role is for an engineering leader who can combine technical depth, strong people leadership, execution discipline, and production judgment. The manager will lead a team delivering complex systems in close partnership with Applied Science, Product, Policy, Security, Privacy, Compliance, and partner engineering teams.

The larger platform scope includes:

• AI-powered Trust & Safety intelligence for understanding advertisers, domains, websites, landing pages, business identities, reputation, provenance, and policy/fraud risk.

• Deep Research Agents and AI-assisted workflows for advertiser investigation, domain review, evidence gathering, policy reasoning, reviewer assistance, appeals, and fraud analysis.

• Decisioning and enforcement systems for policy enforcement, fraud detection, abuse prevention, advertiser risk scoring, and marketplace protection.

• Model-serving and orchestration infrastructure for LLMs, SLMs, wide & deep models, ensembles, classical ML, heuristics, rules, and policy controls.

• Security, Risk, and Fraud systems for threats such as phishing, malware, cloaking, account takeover, payment abuse, fake business identities, compromised advertisers, coordinated abuse, and policy evasion.

• Human-in-the-loop, audit, and governance systems with explainability, reproducibility, reviewability, and compliance readiness.