Junior Threat Context Analyst

Microsoft is seeking a Threat Context Analyst to join the Threat Analysis Center (MTAC) to focus on identifying and mitigating AI abuses. The role involves tracking sophisticated threat actors and writing threat intelligence reports, while contributing to MTAC's mission of detecting and disrupting digital threats. Responsibilities Research and assess cyber and malign influence threats at tactical and strategic levels by drawing on information from social media accounts and websites, foreign policy priorities and perspectives from open-source reporting Understand the components of generative AI and how technology stacks produce AI outputs Identify and triage AI abuses based on behavioral and technical indicators Write threat intelligence reports for audiences on adversary influence actors, networks, and operations powered by artificial intelligence (AI) Work closely with the broader Microsoft Threat Intelligence team in its investigations of nation state cyber, influence, and AI-first actor investigations Develop engaging presentations and brief various stakeholders under tight deadlines Follow innovative, non-intrusive, law-abiding methods for detecting, diagnosing, and deterring the most advanced and prolific threats in the information environment Qualification Data wrangling Python SQL Predictive modeling Generative AI knowledge LLMs experience AI prompt engineering Analytical skills Communication skills Required Master's Degree in Mathematics, Analytics, Engineering, Computer Science, Marketing, Business, Economics or related field AND 1+ year(s) experience investigating a combination of cyber and influence operations in either the public sector, private sector or a combination of the two OR Bachelor's Degree in Statistics, Finance, Mathematics, Analytics, Engineering, Computer Science, Marketing, Business, Economics or related field AND 2+ years experience investigating a combination of cyber and influence operations in either the public sector, private sector or a combination of the two Proficiency in data wrangling and processing pipelines using tools such as pandas, NumPy, or PySpark and fluency in Python and SQL tools with experience collecting large-scale data, building analytics/automation pipelines, creating dashboards, and communicating data-driven insights to non-technical partners Experience building exploratory predictive models, including rapid profiling of new data sets and application of lightweight ML/statistics (clustering, anomaly detection, baseline classifiers/regressors) to surface trends and generate actionable insights and forecasts Preferred Deep understanding of how generative AI systems work end-to-end, from data collection and pretraining through fine-tuning and inference Extensive hands-on use of LLMs and experience red-teaming or evaluating failure modes (jailbreaks, hallucinations, bias, prompt injection, leakage) Understanding of AI prompt engineering and prompt attack methods (e.g., prompt injection, indirect prompt leakage, prompt leaking across contexts) Ability to distinguish different threat actor techniques for employing generative AI and how they can be correlated to attribute threat actors Software, cloud computing, consumer electronics, and enterprise systems. Data wrangling Python SQL Predictive modeling Generative AI knowledge #J-18808-Ljbffr Jobright.ai