Mukul Dayal

Senior Data Scientist @Instnt

NY, US

As a Senior Data Scientist at Instnt, I lead the development of machine learning models for customer identity verification and fraud detection within the financial and identity verification (IDV) domain. My responsibilities include designing end-to-end data science pipelines — from raw data ingestion and preprocessing using AWS Glue and PySpark to advanced feature engineering and model training with tools such as H2O, Scikit-Learn, and Keras.I work closely with cross-functional teams to modernize and scale legacy decision systems by integrating cloud-native and serverless technologies like AWS Lambda. I actively manage model evaluation using metrics like AUC and AUCPR, perform hyperparameter tuning, and ensure robust model deployment and version control using MLflow. Additionally, I contribute to architecture decisions for scalable fraud detection workflows and support real-time inference systems in production.

Meerut Institute of Engineering and Technology(MIET)

Bachelor's degree