Microsoft Security is looking to protect its customers and services from abuse and fraud by developing innovative, intelligent, and scalable risk solutions.
Requirements
- 1+ year(s) experience with machine learning, predictive analytics, or research
- Familiarity with device fingerprinting, fuzzy ID techniques, or entity resolution
- Experience in risk scoring, bot mitigation, or anomaly detection systems
- Proficiency with cloud platforms such as Azure, and real-time inference systems
- Knowledge of statistics, econometrics, computer science, electrical or computer engineering, or related field
- Experience with querying, visualization, and reporting techniques
- Experience with model development and evaluation
Responsibilities
- Develop state-of-the-art machine learning solutions that power real-time fraud and abuse detection and decision-making
- Understand where to acquire the data necessary for your project plan and use querying, visualization, and reporting techniques to describe that data
- Understand modeling techniques, select the correct tool and approach to complete objectives, and evaluate the output for statistical and business significance
- Analyze model performance and incorporate customer feedback into its evaluation
- Write code for a specific feature, and develop a working expertise of proper debugging techniques
- Understand device fingerprinting techniques, analyze the device attributes collected for device fingerprinting, identify anomalous attributes within sessions as well as across sessions
- Develop techniques to devise a unique device identifier given the device attributes
Other
- Bachelor's Degree in Statistics, Econometrics, Computer Science, Electrical or Computer Engineering, or related field AND 2+ years related experience
- 1+ year(s) customer-facing, project-delivery experience, professional services, and/or consulting experience
- Ability to meet Microsoft, customer and/or government security screening requirements
- Communication and collaboration skills, with the ability to influence across disciplines
- Proven ability to optimize detection systems for both technical and business metrics
