Rajvardhan Oak

DevOps Engineer

Seattle, Washington, United States6 yrs experience

Highly Stable

Key Highlights

Expert in machine learning and fraud detection.
Designed end-to-end ML pipelines for large-scale applications.
Strong background in data analysis and security.

Stackforce AI infers this person is a Machine Learning Expert specializing in Fraud Detection within the AdTech and Fintech industries.

Contact

rajoak1995@gmail.com LinkedIn

Skills

Core Skills

Machine LearningFraud Detection

Other Skills

Angular 4Anomaly DetectionCC#C++Computer VisionCyber-securityData AnalysisData MiningDeep LearningFeature EngineeringGraph Neural NetworksJavaKerasLarge Language Models (LLM)

About

I’m a machine learning researcher and applied scientist with a strong focus on Trust & Safety, fraud detection, and online integrity. At Microsoft, I design and deploy ML and graph-based models to combat large-scale advertising fraud, leveraging behavioral signals from web activity, clickstreams, and traffic patterns. Beyond industry, my research has explored fake reviews, harmful content moderation, and malware detection—blending statistical modeling, qualitative insights, and cutting-edge deep learning. I enjoy building systems that sit at the intersection of security, behavior, and data.

Experience

Microsoft

2 roles

Applied Scientist II

Promoted

Oct 2022 – Apr 2025 · 2 yrs 6 mos · Redmond, Washington, United States

I work in Ads Fraud Detection for Microsoft Ads. My work involves analyzing large datasets from multiple sources (click logs, cookies, IP patterns, and more) to detect fraudulent activity like scrapers, click-bots, conversion fraud, etc. I own end-to-end pipelines (feature engineering, training, deploying and monitoring) for various machine learning models (GNN, anomaly detectors, deep neural networks, and random forests) to detect and mitigate large-scale advertising fraud.

Machine LearningFraud DetectionData AnalysisFeature EngineeringDeep LearningGraph Neural Networks+2

Applied Scientist

Jun 2020 – Sep 2022 · 2 yrs 3 mos · Redmond, Washington, United States

Ibm

Data Scientist Intern

May 2019 – Aug 2019 · 3 mos · New York

I worked on internal fraud detection initiatives, with a focus on expense and reimbursement fraud. I designed and deployed machine learning classifiers to flag anomalous claims by analyzing both structured metadata and unstructured OCR outputs from receipts. My models leveraged features from spending patterns, merchant data, and visual/text inconsistencies, using algorithms like XGBoost, Random Forests, and Logistic Regression. I also incorporated NLP techniques such as named entity recognition and fuzzy matching to validate receipt contents against declared expenses. To support operational use, I collaborated with audit and compliance teams and created dashboards in Power BI for ongoing monitoring.

Machine LearningFraud DetectionNLPData AnalysisXGBoostRandom Forests+2