Headline

Data Scientist | Credit Risk Modelling & ML Engineering | AWS Cloud | Statistical Modelling | Predictive Analytics | Financial Services | Immediate Joiner

About

Data Scientist with 3+ years of experience in financial services, combining credit risk modelling with hands-on cloud engineering to deliver end-to-end analytical solutions. The professional focus extends beyond model development to the technology, infrastructure, and operational environment required to make data science solutions practical and scalable. A strong interest lies in solving complex analytical problems through a combination of statistical thinking, machine learning, cloud technologies, and structured problem-solving. Experience spans credit risk initiatives from feature selection and model development through validation, reason-code generation, and production reporting, alongside ownership of cloud infrastructure and optimisation. Core Expertise: - Data Science: Machine Learning | Credit Risk Modelling | Statistical Analysis | Predictive Modelling | Feature Engineering | EDA | Model Validation | Scorecard Development - Cloud & Engineering: AWS | SageMaker | EMR/Spark | Glue | Lambda | Terraform | IAM/KMS | Bedrock | SQL | PySpark | Docker | Jenkins CI/CD - Analytical Strengths: Data Analysis | Problem Solving | Data-Driven Decision Making | Model Explainability | Production Reporting | Business & Technical Collaboration Key achievements: - Credit Risk Modelling: Developed and evaluated statistical and machine learning models for credit risk scoring across diverse business segments and datasets. - Model Development & Explainability: Built feature selection, scorecard, Weight of Evidence, validation, and reason-code methodologies to support robust and interpretable modelling. - Cloud & Data Optimisation: Improved AWS-based analytical workloads through resource optimisation, data architecture enhancements, and workload automation. - FinOps & Cost Optimisation: Developed cloud cost attribution and reporting solutions to improve visibility into spending and identify optimisation opportunities. - Infrastructure Automation: Built and managed AWS infrastructure using infrastructure-as-code, access controls, encryption, and event-driven automation. - Analytical Productivity: Developed reusable data profiling and exploratory analysis solutions to streamline analytical workflows and reduce repetitive effort. Currently exploring Data Scientist, Credit Risk Analyst and ML Engineer roles in Bengaluru or remote Open to connecting with professionals and teams working at the intersection of Data Science, Machine Learning, Financial Services, and Cloud Engineering.

Location

Delhi

AI stack

claude

Work experience

Analytic Science -Associate at FICO, Jul 2024 - Jul 2026 · 2 yrs 1 mo, CREDIT RISK MODELING • Developed and evaluated logistic regression and neural network models for credit risk scoring across multiple datasets, including sparse and explicit sparsity variants, achieving up to 0.94 AUC • Engineered a feature selection pipeline (Fisher's score, mutual information, LASSO, random forest) with Optuna hyperparameter tuning, reducing feature count 48% while preserving model performance • Applied WOE binning, scorecard development, validation, and reason code generation to build interpretable, explainable credit risk models • Used supervised and unsupervised methods across classification, regression, clustering, and forecasting problems MODEL AUTHORING PLATFORM • Deployed a SageMaker Studio domain hosting 6 self service analytical templates , each scoped to its own AWS resources via per template IAM policies • Validated access controls across multiple AWS accounts including production, confirming users could reach only the tools they had requested FALCON INDIA • Stood up a new environment end to end: provisioned EC2 and EFS with Terraform, defined IAM roles, trust policies, and KMS encryption • Automated instance scheduling with Lambda, EventBridge and SNS, cutting idle EC2/EFS costs an estimated 42% • Optimized EMR and Glue Data Catalog workloads through root cause bottleneck analysis bringing CPU utilization down 71%, memory usage down 62% • Migrated Hive tables to Iceberg via Athena, accelerating SQL extraction and updates (CPU : 23%, memory : 26%) FINOPS & ANALYTICAL TOOLING •Automated AWS cost reporting via Lambda, Cost Explorer, and SES, using tag-based attribution to surface $400/week in unused service spend resulting in shut down on discovery, ~$21K annualized • Prototyped a unified profiling, distribution and correlation analysis tool, cutting repeated EDA setup an estimated 70% across 4 datasets.

Data Science Intern at FICO, May 2023 - Jul 2024 · 1 yr 3 mos, 14-month internship alongside final year B.Tech studies; converted to full time Associate. • Benchmarked and unit tested internal data analysis tools, improving processing time 38% on datasets of 2.3 million rows • Identified and engineered features that improved model accuracy, robustness, and predictive performance • Designed and analyzed experiments to test product, process, and business hypotheses, translating results into recommendations for the analytics team • Built working familiarity with the credit risk modeling lifecycle data preparation, EDA, and validation that carried directly into the Associate role

Business Consultant at EY, May 2022 - Jul 2022 · 3 mos, Gurugram, Two month undergraduate internship. Analyzed client operational data and built Power BI and Excel reports that managers used in client facing work. First hands on exposure to delivering analysis in a professional setting.

Education

Bachelor of Technology - BTech, Computer Science, Kalinga Institute of Industrial Technology, Bhubaneswar

Delhi public school