Scribd
Upload
140 views1 page

Varun Surapaneni: Experience

Varun Surapaneni has experience as a Credit Strategy Analyst at WebBank and as an Equity Trading Intern at Loop Capital. He has a Master's degree in Operations Research from Columbia University and a Bachelor's degree in Industrial Engineering from IIT Delhi. His skills include finance, analytics, programming languages like Python, SAS, and Tableau. At WebBank, he developed models to forecast loan losses and monitor credit risk. At Loop Capital, he enhanced algorithmic trading strategies and built backtesting tools.

Uploaded by

Purshottam Kandhari
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
140 views1 page

Varun Surapaneni: Experience

Varun Surapaneni has experience as a Credit Strategy Analyst at WebBank and as an Equity Trading Intern at Loop Capital. He has a Master's degree in Operations Research from Columbia University and a Bachelor's degree in Industrial Engineering from IIT Delhi. His skills include finance, analytics, programming languages like Python, SAS, and Tableau. At WebBank, he developed models to forecast loan losses and monitor credit risk. At Loop Capital, he enhanced algorithmic trading strategies and built backtesting tools.

Uploaded by

Purshottam Kandhari
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

310 10th St, Apt 206 (917) 678-5024

Jersey City, NJ 07302 VARUN SURAPANENI [email protected]


EXPERIENCE
WebBank, New York Credit Strategy Analyst Feb 19 – Apr 20
Credit risk management of Consumer and SMB lending portfolios; evaluation of new opportunities in Fintech lending space
Modelling:
• Modelled loss and paydown curves for consumer lending partners like Avant, Lending Club, Aura (Insikt) and Mosaic

• Summarized loan level data in SAS; Developed loan loss forecasts in Python using regression (Linear, Polynomial, Quantile)

• Removed outliers with ellipsoid fitting; Back tested using MAPE, MSE and monitored model performance on a quarterly basis

• Developed loan-level loss forecasting model to using Cox PH model to observe how losses vary across different vintages

• Quantified contractual/non-contractual payments using CPR (Cumulative Prepayment Rate) to forecast portfolio paydowns

• Reviewed ML models; internal ERM and external underwriting, fraud, loss forecasting models to explain portfolio performance

Portfolio Monitoring:
• Computed portfolio level risk adjusted margins on a monthly basis and explained trends/variations to the executive team

• Monitored consistency in loan origination metrics across Bank’s retained portfolio and overall platform using Wilcoxon, t-tests

• Presented KPI reports on Toast Capital loans to executive team to better understand Bank’s economics and loss protections

• Identified synthetic id fraud pattern related to 220 loans of $3MM in total size; among loans of a major consumer lending partner

• Automated daily reporting of COVID-19 related loan hardships/charge-offs using SAS as a database management tool

Financial Planning and Analysis:


• Actively involved in Bank’s annual budgeting processes; Estimated originations for next financial year from partner policy changes

• Presented portfolio specific scenario analysis showing impact on net income to executive team to drive decision making

• Provided quarterly updates to ensure loan loss reserve & budgeting processes are tied to forecasts

Cross-team Collaboration:
• Provided performance updates and comparable company analysis to Business Development team to assist in future prospects

• Supported Capital Markets team as and when required with reports on portfolios involved in secondary market deals

• Assisted Bank's Finance team in booking monthly entries and reconciling with loan level data received from counter parties

• Worked with Data and MRM teams, to sort out data anomalies while transitioning to Lending Club new investor reporting

Loop Capital, New York Equity Trading Intern June 18 – Sep 18


Real-time analytics support on high value quantitative technology projects; including supporting the algorithmic trading business
• Developed intraday execution strategy in Python using technical indicators such as Moving Average and Relative Strength Index
• Enhanced accuracy of Volume Imbalance strategy using SVM on tick data; Predicted stock price direction and generated signals
• Built an inhouse back-testing framework for efficiency; Tested strategies on a universe of 5000 tickers for last 6-month data
• Provided a real time analytics tool that tracks technical indicators like Bollinger bands, Stochastic Oscillator etc. at a ticker level
SKILLS
• Finance and Analytics: Loss/Paydown forecasting, ALLL estimation, Stress testing, FP&A, Data Visualization, Web Scraping
• Computing: Python (NumPy, Pandas, Matplotlib, Scikit-Learn, TensorFlow, Keras), C, SAS, SQL, Tableau, VBA, MS Excel
EDUCATION
Columbia University New York, US Aug 17 – Dec 18
• Master of Science in Operations Research; Concentration: Business Analytics, Financial & Managerial Applications GPA: 3.8/4
• Coursework: Machine Learning, Time Series, Model Based Trading, Structured and Hybrid Products, Fundamentals of Financial

Engineering, Programming for Financial Engineering, Data Analytics, Stochastic Modelling, Optimization, Probability and Statistics
Indian Institute of Technology Delhi New Delhi, IN Jul 12 – May 17
• B.Tech in Industrial Engineering with Minor in Business Management (1% acceptance rate among 500K applicants) GPA: 7.5/10
• Relevant Coursework: Financial Mathematics, Banking & Financial Services, Managerial Accounting, Calculus, Data Structures

ACADEMIC PROJECTS
Deep Learning in financial time series prediction, Columbia University Jan 18 – May 18
• Identified high-level features in past 8-year daily stock price data of Indian index NIFTY using Stacked Autoencoders (SAE)

• Incorporated these features into Long short-term memory (LSTM) network in Python Keras to forecast next day’s closing price

• Generated buy-sell signals using LSTM forecasts; strategy resulted in return of 9.5% over 3-month period with MAPE of 2.6%

Russell 1000 Large-Scale Portfolio Optimization, Columbia University Aug 17 – Dec 17


• Replicated portfolio for Russell 1000 with few dominant stocks by applying factor models using historical stock price in Python

• Calculated factors using Jacobi method to do Principle Component Analysis; thereby obtained the respective residual returns

You might also like