Mitul Shah

Data Scientist

Vadodara, Gujarat, India12 yrs 5 mos experience

Most Likely To SwitchAI ML Practitioner

Key Highlights

Built predictive models with high accuracy for educational institutions.
Led data-driven initiatives that improved business performance significantly.
Expert in machine learning and data analytics for actionable insights.

Stackforce AI infers this person is a Data Science expert with a strong focus on predictive analytics in Education and Telecommunications.

Contact

shahmitul1809@gmail.com LinkedIn

Skills

Core Skills

Machine LearningData AnalysisAnalyticsData EngineeringBusiness AnalysisData Science

Other Skills

A/B TestingArtificial IntelligenceBayesian statisticsComputer VisionData MiningData VisualizationData WarehousingDeep LearningExploratory Data AnalysisGitGithubGraph TheoryHadoopHierarchical ClusteringHypothesis Testing

About

I am a Data Scientist with expertise in leveraging machine learning, statistical modeling, and data analytics to uncover actionable insights. My work focuses on building predictive models and algorithms that solve complex business challenges, enabling organizations to make data-driven decisions with confidence. I am passionate about solving real-world problems using data. I enjoy exploring large datasets, identifying patterns, and creating solutions that enhance business performance. Whether it’s optimizing marketing campaigns, improving product offerings, or forecasting trends, I thrive on transforming data into valuable outcomes. Proficient in Python, R, SQL, and machine learning frameworks like TensorFlow, Scikit-learn, and Keras. Extensive experience in building predictive models, clustering, classification, and regression analysis. Expertise in natural language processing (NLP), deep learning, and big data technologies. Ability to collaborate with cross-functional teams to implement data-driven solutions that align with business goals. I am committed to continuous learning and staying at the forefront of data science innovations while applying my skills to create meaningful impact. I can be reached on shahmitul1809@gmail.com.

Experience

12 yrs 5 mos

Total Experience

1 yr 9 mos

Average Tenure

2 yrs 7 mos

Current Experience

Shital academy

Senior Data Scientist

Nov 2023 – Present · 2 yrs 7 mos · Vadodara, Gujarat, India · On-site

Built predictive models to predict total student enrollment with an accuracy of 95%, enabling better faculty hiring and resource
planning.
Developed predictive models that accurately detected struggling students early (85% recall), helping the institution take timely
action and improving student retention by 12%.
Conducted A/B testing and statistical analyses to evaluate new teaching formats on the website, which improved student satisfaction scores by 10% and led the institution to adopt the new approach.

Python (Programming Language)Exploratory Data AnalysisA/B TestingAnalyticsMachine LearningData Analysis

Aivhub ltd

Senior Data Engineer

May 2023 – Nov 2023 · 6 mos · Remote

Developed time-series forecasting models predicting network traffic with 90% accuracy, resulting in better planning of network resources and fewer chances of the network slowing down for British Telecom.
Designed customer segmentation models using K-Means and hierarchical clustering, enabling targeted cross-sell campaigns for British Telecom that promoted the most relevant services to each segment, such as offering higher-speed broadband to heavy internet users.

R ProgrammingA/B Testingk-means clusteringHierarchical ClusteringBusiness AnalysisData Engineering

Bookmyshow

Data Science Manager

Nov 2021 – Sep 2022 · 10 mos · Mumbai, Maharashtra, India · Hybrid

Built movie recommender system for BookMyShow platform, increasing click-through rate (CTR) by 18%.
Led a 5-member team to deploy A/B testing frameworks optimizing homepage UI, achieving a 7% increase in click-through rates.
Increased campaign ROI by 18% by building personalized customer segmentation models that grouped users by viewing patterns, purchase history, and engagement, enabling targeted offers such as movie discounts for movie-focused users and sporting event offers for sports-interested users.

R ProgrammingPythonRecommender SystemsA/B TestingBusiness AnalysisData Science

Cars24

Data Science Manager

May 2021 – Nov 2021 · 6 mos · Gurugram, Haryana, India · On-site

Launched 0% down payment feature after A/B testing on 100,000+ customers, boosting conversion by 14%.
Built dynamic interest rate models by first predicting each customer’s default probability and then assigning risk-based interest rates, increasing loan approvals by 11% while keeping default levels under control.
Clustered vehicles to identify high-demand categories, improving loan disbursement by 8%.

PythonBusiness AnalysisR ProgrammingA/B TestingData AnalysisData Science

Etech global services

Data Analyst

Dec 2020 – May 2021 · 5 mos · Remote

Built linear regression models to predict overall customer experience for Fortune 500 clients using multiple customer behavior and engagement variables.
Generated actionable, data-driven insights highlighting the key variables impacting customer retention, enabling clients to prioritize initiatives effectively.
Presented findings to C-level stakeholders, leading to strategic investments of $2M+ to improve targeted areas and enhance customer retention.

Mueller labs

Data Scientist

Jan 2015 – Nov 2020 · 5 yrs 10 mos · Houghton, Michigan, United States · On-site

Developed a machine learning–based churn prediction system for the gaming platform that identified high-risk players and enabled personalized offers, messages, or rewards to encourage them to stay active, reducing player attrition by 15%.
Increased revenue by 200% by identifying low-revenue segments and launching targeted ‘two-for one’ promotions validated through A/B testing for a gaming company.
Extended R pathmapping library for applications in education and behavioral science and co-authored white papers presented at 2 academic conferences.
Built credit card fraud detection models achieving 96% recall, ensuring that the system successfully identified nearly all fraudulent transactions with minimal misses.
Used clustering methods to uncover anomalous transaction patterns, helping reduce the fraud investigation backlog by 20%.