Deepak Bhatt Ph.D

Director of Engineering

O'Fallon, Missouri, United States15 yrs 3 mos experience

Most Likely To SwitchAI Enabled

Key Highlights

Expert in developing advanced fraud detection models.
Proven track record in leading data science teams.
Strong background in demand forecasting and time series analysis.

Stackforce AI infers this person is a Fintech expert specializing in Machine Learning and Data Science.

Contact

deep_83bhatt@yahoo.co.in LinkedIn

Skills

Core Skills

Machine LearningFraud DetectionData ScienceDeep LearningDemand ForecastingTime Series AnalysisMulti-sensor Data Fusion

Other Skills

AlgorithmsArtificial IntelligenceAssistant TeachingCC++Decision IntelligenceEnvironmental modelingFeature DiscoveryFeature SelectionIMU Error CharacterizationJavaLSTMLaTeXLabVIEWMathematical Modeling

About

Advancing Machine Learning approaches to build best in class payment fraud detection models powering Decision Intelligence (DI) Experience spans from discovering insights from evolving fraudulent patterns, building robust and scalable Machine Learning models and performance testing to deliver high performant ML models KPIs in production. Past experience includes grooming and building a team of data scientists to solve business problems across various sectors such as forecasting, payment fraud using data driven decision making

Experience

Mastercard

2 roles

Director of Data Science

Promoted

Jan 2022 – Present · 4 yrs 2 mos

Leading the development of DI Pro to advance our fraud detection capabilities. Led the advancement of Machine Learning modeling capabilities to build best-in-class Decision Intelligence (DI) fraud detection capabilities.
Leading the development of payment fraud detection models to reduce fraud in the Mastercard Send ecosystem
Developed scalable HPO capabilities to tune fraud detection models, led the advancement of feature discovery and feature selection. This advances our modeling capabilities that delivers stable and high performant fraud detection models

Machine LearningFraud DetectionDecision IntelligenceFeature DiscoveryFeature Selection

Data Science Manager

Nov 2019 – Dec 2021 · 2 yrs 1 mo

Leading team of Data Scientists to drive efficiency using data driven insights/modeling to solve various business challenges such as churn/retention, Life time value (LTV), risk modeling. Trend modeling across cross border transactions using template matching on time series data.
Advancing state of the art deep learning methods to solve for privacy, model bias/fairness, synthetic data, adversarial robustness and etc.

Data ScienceDeep LearningRisk ModelingTrend ModelingTemplate Matching

Koch technology center (ktc)

Data Scientist

Aug 2018 – Nov 2019 · 1 yr 3 mos · Greater Bengaluru Area

Building demand forecasting models using internal and external indicators (such as historical sales, market trends etc).
Collaborating with various Koch industries: Flint Hills Resource (polypropylene demand), Invista (Synthetic Dyed Nylon) and Guardian (Glass demand in Europe) to model their demand appropriately thereby helping them take informed decisions and meeting timely customer demand.
To model thousands of products jointly, LSTM based approach and wavenet is implemented for improved forecasting accuracy against traditional baseline.
R&D Efforts:
Developing time series clustering method to reduce the scale of external factors possessing redundant signals.

Demand ForecastingTime Series AnalysisLSTMWavenet

American express

Manager- Machine Learning/Data Science

May 2017 – Aug 2018 · 1 yr 3 mos

Wipro

Technical Lead-Data Science

Jul 2016 – May 2017 · 10 mos

Samsung electronics

2 roles

Technical Lead-Research

Mar 2016 – Jul 2016 · 4 mos

Lead Engineer-Research

Feb 2014 – Feb 2016 · 2 yrs

Merl

Research Intern

May 2013 – Aug 2013 · 3 mos · Greater Boston

The university of toledo

Graduate Research Assistant

Aug 2010 – Dec 2013 · 3 yrs 4 mos

Research involving mutli-sensor data fusion methodologies for positioning applications and IMU (Inertial Measurement Unit) error characterization.
Proctored undergraduate level labs.
Electric Circuits lab-2300, (Summer 2012)
Electronic Design, (Spring 2012)
Engineering Technology, (Summer 2011)
Electric Circuits lab-2300, (Spring 2011)
Digital Logic Design lab, (Fall 2011)
Electronics lab-3400, (Fall-2010, Spring 2011)

Multi-sensor Data FusionIMU Error Characterization