Data Analyst
Posted Updated
Description
Responsibilities:
- Design, develop, and maintain interactive dashboards and reports using Tableau, Power BI, or similar BI tools to support business stakeholders.
- Write complex SQL queries to extract, transform, validate, and reconcile data from relational databases and data warehouses.
- Develop Python-based automation scripts for repetitive reporting, data processing, data quality checks, and reconciliation workflows.
- Perform exploratory data analysis to identify trends, patterns, outliers, and business drivers across large BFSI datasets.
- Build and validate basic-to-intermediate data science models such as classification, regression, forecasting, segmentation, propensity scoring, and anomaly detection models.
- Translate business problems into analytical frameworks, define target variables, engineer features, evaluate model performance, and explain model outputs in business-friendly language.
- Work on use cases such as delinquency/bounce prediction, collections strategy, branch performance risk flags, portfolio yield analysis, and operational exception detection.
- Create model monitoring and performance tracking reports to measure lift, stability, drift, accuracy, recall, precision, and business impact over time.
- Collaborate with business, product, operations, risk, collections, finance, and technology teams to convert insights into actionable interventions.
- Track analytics tasks, sprints, documentation, and knowledge repositories using JIRA and Confluence.
- Present analytical findings and recommendations clearly to both technical and non-technical audiences, including senior leadership.
Requirements:
- SQL Must-Have: Advanced querying, joins, CTEs, window functions, stored procedures, data validation, and query optimisation.
- Python + Git: Must have Pandas, NumPy, automation scripting, data wrangling, reusable code, version control, and basic packaging discipline.
- Data Science / ML Must-Have / Strong.
Good-to-Have:
- Scikit-learn, statsmodels, regression, classification, clustering, time-series forecasting, feature engineering, train-test validation, cross-validation, confusion matrix, ROC-AUC, precision, recall, lift, and model explainability basics.
- Statistics Must Have: Hypothesis testing, correlation, distributions, sampling, outlier treatment, cohort analysis, trend analysis, and variance/driver analysis.
- AWS Redshift Must-Have: Data warehousing, cluster usage awareness, query performance tuning, and working with large production datasets.
- Automation and ETL: Must-have scripted pipelines, scheduled jobs, reconciliation checks, exception reporting, and ETL automation.
- BI / Dashboarding Must Have Tableau / Power BI dashboard design, KPI tracking, drill-down views, stakeholder-ready reporting, and report publishing.
- Good-to-Have: Ability to convert models into repeatable scoring scripts/batch jobs, document assumptions, and coordinate with engineering teams for productionization.
- Documentation: Must Have Clear documentation of logic, assumptions, data definitions, model methodology, limitations, and business interpretation using Confluence or similar tools.
Non-Technical Skills:
- Problem Solving: Ability to break down complex business problems and design data-driven solutions with minimal supervision.
- Business Translation: Ability to convert model outputs into simple business actions, not just technical metrics.
- Stakeholder Management: Proven ability to communicate insights clearly to diverse audiences, including senior leadership, product teams, operations, risk, and collections.
- Ownership Mindset: Strong follow-through on data accuracy, timelines, documentation, and stakeholder closure.
- Communication: Strong written and verbal communication skills with the ability to simplify complex analysis.
- Agility: Comfortable working in fast-paced environments and managing multiple priorities simultaneously.
Preferred Background:
- 1-5 years of experience in analytics, business intelligence, or data science roles, preferably in BFSI / NBFC / lending/fintech.
- Experience working with loan, customer, collections, risk, operations, sales, finance, or portfolio datasets will be an advantage.
- Exposure to production reporting environments and large-scale data warehouses will be preferred.
- The candidate should help the organisation move beyond static MIS reporting towards insight-led analytics, automated decision support, and practical data science models that improve business monitoring, prioritisation, risk identification, productivity, and leadership decision-making.
Skills
- Analytics
- Anomaly Detection
- Automation
- AWS
- Business Intelligence
- Confluence
- Data Quality
- Data Science
- Data Warehousing
- ETL
- Feature Engineering
- Fintech
- Git
- Jira
- Machine Learning
- NumPy
- pandas
- Power BI
- Python
- RDBMS
- Redshift
- scikit-learn
- SQL
- Stakeholder Management
- Statistics
- Tableau
- Time Series
- Version Control
Sign in to read and join the discussion.