About me

Skilled professional with over 2.5 years of experience in Data Science and Analytics within the Fintech industry. Proficient in Technical, Product, and Business aspects, including Product and Risk Analytics, Business Intelligence, Data Modelling and ETL Processes, Data Science, and Data Visualization.

Currently, I'm working as a Data Analyst at Juno Finance based out of Bengaluru, India.

What I do?

  • design icon

    Product and Business Analytics

    Analytics, Deep-Dives and Hypothesis Testing for finding meaningful insights.

  • Web development icon

    Data Engineering

    Data-pipeline Developement from setting up ETL processes, Data Modelling to workflow orchestraction.

  • mobile app icon

    Data Science

    Analytics / Machine-Learning Models for Fraud Detection / User Quailty / Customer Segmentation / Text Analytics.

  • camera icon

    Data Visualization

    Presenting Insights in a concise and easily understandable way.

Resume

Experience

  1. Data Analyst
    Juno Finance

    Oct 2023 — Present

    - Collaborated with Product, Business, and Risk teams to conduct comprehensive data analysis pertaining to Disputes, Fraud Detection and other key aspects which resulted in corrective actions, such as reduced disputes losses by more than 40%, as well as increased net-revenue.
    - Achieved a substantial reduction in CRM data retrieval latency (to 1/4th of the original time) through effective partitioning and indexing methods. Instrumental in implementing a smart-search functionality within the CRM system, significantly accelerating data analysis processes.
    - Developed a real-time user-level dataset, with around 200+ variables (Risk + Product + CS) helpful in identifying fraudulent users and segmenting user cohorts, while at the same time helping the support team enhance customer assistance and satisfaction.
    - Engineered complex ETL scripts to clean, transform, and integrate diverse datasets, elevating data quality and enabling informed business decisions. Implemented efficient data modeling techniques and optimized job scheduling to strategically reduce costs by 50% and enhance system performance.
    - Automated labor-intensive tasks, such as extracting user data from unstructured data, which would otherwise demand multiple days of manual-human effort. Similarly, addressed issue related to understanding the reasons pertaining to negative balance, a process that traditionally consumed a considerable amount of manual hours.

  2. Data Analyst
    Jupiter

    Dec 2021 — Sept 2023

    - Built the analytics pipeline from 0 to 1 for Rewards, Referral and Gift cards - from datawarehousing in dbt to Final Dashboards on Metabase and Tableau - resulting in faster execution of queries and report generation time by around 60%.
    - Helped increase the referral run-rate from 450 to 850 accounts/day by working with the referral team to identify key-actions which leads to increased referral engagement and heavy experimentation around the same.
    - Analysed the user's discovery and interaction with major products on Homepage, to figure out refinement areas, in order to improve Traffic, CTR, Conversion as well as reducing Contacts per Customer (CPC).
    - Discovered multiple critical bugs in the rewards disbursal logic which were leading to Rewards being abused, thereby reducing company's burn by around 1.5 lakh/day.
    - Examined the mapping accuracy of the Rewarded refunds and reversal Transactions resulting in 13% increase in rewards clawback MoM and 2x improved Spends Insights for these transactions.

  3. Research Intern
    University of Twente

    Nov 2020 - April 2021

    - Removing the data gaps in the MODIS Terra FSC dataset using a two step process: Spatial and Temporal Filtering and comparing it with observation station data.
    - Analyzing trends in the snow cover using MK test and correlating it with precipitation and temperature data.

  4. Research Intern
    The University of Tokyo

    June 2019 - July 2019

    - Found and removed discrepancies such as void regions and incorrect coastal boundaries present in TanDEM-X data.
    - Performed Comparison with datasets such as ASTER, AW3D and Error component analysis using ICESat data.

Education

  1. The University of Tokyo (Dropped Out)

    2021

    M.Sc. in Civil Engineering (Hydrology and Climate Change)

  2. Indian Institute of Technology Roorkee

    2016 - 2020

    B.Tech. in Civil Engineering (CPI: 8.91/10)

Languages & Tools

  • SQL, Python
    80%
  • Data Visualization Tools, MS-Office
    70%
  • Redshift, Athena, Bigquery
    85%
  • Hevo, dbT, git, bash, HTML / CSS
    65%
  • Airflow, Kafka, Spark, Docker
    50%

Blog

Contact

Contact Form