Feb 2025 – Jun 2025 · 4 mos · Bengaluru, Karnataka, India

• Took a break to spend time with my child and family

Aug 2022 – Jan 2025 · 2 yrs 5 mos · Bengaluru, Karnataka, India · On-site

• Nira provides instant personal loans to middle income segment in India. Working with Data Sci-
• ence, Credit, Tech and Growth teams to improve unit economics and manage risk of Personal Loans.
• Alongside, leading a team of 4 data scientists and engineers.
• Increased approval rates by 20% and ticket size by 80% through cohort analysis
• Reduced credit underwriting operational costs by 50% through automation of approvals while
• maintaining low NPAs
• Improved risk management by developing a credit scoring model
• Launched personal loans for Self Employed segment
• Identified product improvements to increase user retention
• Reduced CAC by removing external APIs for Bank Statement analysis by building in-house solution
• Improved transparency of models to business stakeholders by streamlining documentation and
• processes

Dec 2017 – Mar 2022 · 4 yrs 3 mos

• Leading a team of six mortgages structuring strats in Bangalore, responsible for structuring and
• pricing of RMBS and ABS securitizations in EMEA and APAC regions
• Lending Platforms
• Designed different home loan products based on market demand and opportunity
• Analysed the impact of different loan features on advance rates by developing rating agency
• models for securitization of the projected loan portfolio
• Created securitization structures and waterfalls incorporating investor preferences and estimated securitization exit price of different loan products in different economic scenarios
• Back-solved mortgage rates for different loan products to optimise demand/PnL of origination
• Launched lending platforms in Japan, UK and Netherlands and securitized the loan portfolios
• Structured senior financing deals for fintech startups in Europe and India
• Loan Portfolio Pricing
• Priced different European asset backed PL/RPL/SPL portfolios based on rated securitisation Exit
• Developed rating agency models to estimate PD LGD on the portfolio in different rating scenarios
• Created innovative structures and waterfalls to maximize securitisation exit price
• Key contributions
• Improved team efficiency and turn around times by developing automation tools, documentations and creating knowledge sharing processes
• Improved collaboration with different teams and work allocation

Dec 2017 – Dec 2020 · 3 yrs

Aug 2016 – Dec 2017 · 1 yr 4 mos · Bengaluru, Karnataka, India

• Worked on functional and performance improvement of Java based Portfolio Accounting System
• Developed an end to end application to detect potential outliers in portfolio characteristics data

May 2015 – Jul 2015 · 2 mos · Greater Edmonton Metropolitan Area

• The internship was supported by MITACS Globalink Fellowship Grant.
• Worked on two different projects under Prof. Russ Greiner, Department of Computing Science.
• 1. Used supervised learning (Gaussian process regression and Support vector regression) to learn a model to forecast a Type 1 Diabetes patient's Blood Glucose level at the time of next meal given the patient's Diabetes diary. The data is gathered for real patients using a mobile application called EASI-IDM.
• 2. Used supervised and reinforcement learning techniques to best adjust the Type 1 Diabetes patient’s current policy of insulin injection, to improve the long-term "reward" using the data generated by T1DM simulator.

May 2014 – Jul 2014 · 2 mos · Tokyo, Japan

• Worked on functional improvement of auto-testing tools
• Developed a search tool for searching test-cases in an auto-testing tool called JMent
• Developed another search tool for searching files with similar HTML in an auto-testing tool Howl, generated on execution of tests .
• Fixed a number of bugs in JMent
• Worked on java, javascript, sql, used Play framework.

May 2013 – Jun 2013 · 1 mo

• Learnt Web development
• Worked on PHP, MySQL, javascript
• Learnt MVC architecture
• Worked on Ruby on Rails

Apr 2013 – Jun 2013 · 2 mos · IIT Kharagpur

Jul 2011 – Apr 2016 · 4 yrs 9 mos · Kharagpur

• Identification of Intrinsically Disordered Proteins using Machine Learning
• Developed feature set out of attributes of the amino acid sequence for Random Forest classifier
• Developed a baseline using existing disorder prediction programs namely SLIDER, ANCHOR, IUPRED
• Achieved a classification accuracy of 89% for 1539 protein sequences taken from Disprod database
• Promoter prediction using machine learning
• Used supervised learning to identify promoter sequences in the genome of Entamoeba Invadens
• Classified the predicted sequences into encystation specific and non-specific promoters using SVM
• Opinion dynamics in social network
• Developed an algorithm to find minimum number of nodes to be controlled in a network so that
• the network gives a desired final opinion; used Google PageRank algorithm to identify driver nodes
• Developed a linear model for asynchronous opinion dynamics; using regression found the influence
• matrix of the network data gathered from Twitter and Reddit