- 🔭 I just earned these certificates
- 🌱 I’m currently learning everything 😅
- 👯 I’m looking for companies to build their analytics
- 🥅 2022 Goals: To be alive
- ⚡ Fun fact: I'm a polyglot, and I love to play piano
- 🐕 🐈 🎶 ☕ 🍫 🍕 🍗 🌳 🚴
- Used decision tree models to help clients understand their customer demographics
- Based on data from the US Census Bureau, helped clients to understand store-wise performance
- Suggested renewal of old stores as well as setting up of new stores at specific locations to increase sales
- Modelled yearly Goals for 4000+ retail stores which affected the compensation of 100000+ associates
- Critical insights shared with the managers regarding for improving their annual bonus compensations
- The model was generalised for yearly goal settings built on sound statistical and AI driven mechanism
- Created an NPS driver database insights dashboard for the executives to understand customer behaviour
- Variables such as order & store attributes, demographics, web experience, order & pickup experience were taken
- Key factors were found behind the changing NPS scores among categorical store models based on random forest algorithm
- Dealt with 1000+ ad-hoc fraud investigation requests from store managers
- Data cleansing on customer experience backend data based upon suspective IPs and Computer IDs
- Risk reduction of compensatory mechanism by catching and blocking doubtful customers or ex-employees
Product Management for an Indian programming learning Start-Up SOAL
- Researched on various methodologies for modern web-dev frameworks like NodeJS, React & Electron
- Deployed multiple web products in Heroku and AWS and maintained the backend DB in Postgresql
- Developed early learning modules for advanced full-stack web development & data structures and algorithms
Moment Matching Scenario Generation Algorithm for Financial Portfolio Optimisation TPU, Russia
- Worked with Dr. Semenov in the Department of Higher Mathematics and Mathematical Physics at TPU
- Generated scenario trees for multistage problems using the mean CVaR model for portfolio optimization
- Found that a negative probability scenario is generated in the algorithm if random parameters are chosen
- Found that Ponomareva et al, 2015 missed the negative cases, though the required results can be generated
Wavelets and its Applications (DOI: 10.32628/CSEIT183820)
- Worked with Dr. Pravir Dutt in the Department of Mathematics and Statistics, IIT Kanpur
- Researched about image compression algorithm in wavelet analysis based image processing
- Worked with mathematical advancements on Wavelet Analysis and transformation of functions
- Connected wavelet theory with image compression for high definition images and Google Maps application
Mathematical Model of an Ideal Time Table (DOI: 10.13140/RG.2.2.19310.00327)
- Worked with Dr. Prawal Sinha in the Department of Mathematics and Statistics, IIT Kanpur
- Generated scenario trees for multistage problems using the mean CVaR model for portfolio optimization
- Found that a negative probability scenario is generated in the algorithm if random parameters are chosen
- Found that Ponomareva et al, 2015 missed the negative cases, though the required results can be generated
Heuristic Analysis of Growth of Paytm (DOI: 10.17577/IJERTV6IS040337)
- Worked with Dr. Pakrashi in the Department of Economic Sciences, IIT Kanpur
- Conducted primary survey for data collection and studied the impact of PayTM on e-commerce in India
- Studied different factors and economic circumstances responsible for the exponential growth of Paytm
- Worked with Dr. Amit Mitra in the Department of Mathematics and Statistics, IIT Kanpur
- Concluded that for the first 7 years, BTC followed Auto Regressive Moving Average ARMA(7,6) model
- Estimated & Eliminated trend and seasonality in the historic Bitcoin vs USD exchange data of 7 years