Jul 2024 – Jun 2025 · 11 mos · Mumbai, Maharashtra, India · On-site

Sep 2021 – Jul 2024 · 2 yrs 10 mos · Bengaluru, Karnataka, India · On-site

Aug 2020 – Aug 2021 · 1 yr

• Provided custom business insights to strategic US merchants of American Express across travel, lodging, and retail portfolios
• Spearheaded customer acquisition, covid-tracker, and marketing insights for clients leveraging clustering, boosting, and other ML algorithms
• Performed analysis for Ecomm-shift, customer profiling, and competition benchmarking using hive, spark, and other big data tools

Apr 2020 – Jul 2020 · 3 mos

• Designed, implemented and deployed robust algorithmic trading strategies for live trading in US equity and derivative market
• Explored different indicators and back-tested different trading ideas by analyzing stock and options data from interactive brokers
• Implemented and coded strategies using python API from Interactive brokers and used SQL for data basing

Nov 2019 – Feb 2020 · 3 mos · IIT Kanpur

• Successfully conducted a 2-day long Non-Core Weekend with Series of talks, Competitions, and Workshops.
• Sessions ranging from "Carriers in 21st Century", "Algo-trading", "Product management", and "Fintech" by industry experts to "Academia versus Industry" by Academia Professionals.
• Workshops on "Trading Strategies", "Machine Learning" , and "Consulting" are planned to be Organized.
• Competitions on Stock Trading and Business Quiz are scheduled.

Oct 2018 – Mar 2020 · 1 yr 5 mos · IIT Kanpur

• Coordinated a team of 20 members, and managed an annual budget of Rs. 50k, to promote financial literacy on campus
• Organized over 10 + sessions, competitions and workshops to explain and promote financial literacy on campus
• Offered 4 summer and 3 winter projects each for 10 + students in fintech and algorithmic trading domain
• Published monthly and quarterly newsletters to address the appetite of fin and current market trends for campsites
• Planned to organize 2-day non-core weekend with workshops by industry experts and expected a footfall of 1000
• Associated with a fintech startup and bargained a simulation course saving Rs. 15k along with 4 internships
• Administered FB page actively and started the Jargon Made Easy campaign resulting in an increase of 2000 followers

May 2019 – Feb 2020 · 9 mos · Kanpur, Uttar Pradesh, India

• bqror: Bayesian Quantile Regression for Ordinal Models
• Provides an estimation technique for Bayesian quantile regression in ordinal models.
• Two algorithms are considered - one for an ordinal model with three outcomes and the other for an ordinal model with more than 3 outcomes.
• It further provides model performance criteria and trace plots for Markov chain Monte Carlo (MCMC) draws.

May 2019 – Jul 2019 · 2 mos · Chennai Area, India

• Machine Learning and AI intern
• Extracted text from pdf, docx, etc. any format possible using a simple trick. So we first convert every possible format to pdf and then extract text from pdf using PdfMiner.
• Split the text extracted based on different keywords and named entity recognition and created a dataframe to make it look much more structured.
• Applied Named Entity Recognition (NER) for extraction of College name, degree, GPA, personal details, work duration different organizations, skills etc.
• Performed multi-class classification using deep learning to classify resumes as being recommended or not being recommended. Also implemented SVM model based on TF-IDF for classifying the content. Performing TF-IDF and Count Vectorizer as well.
• Performed text summarization using text rank algorithm and using deep learning model based on encoder, decoder LSTM
• Performed exploratory data analysis (EDA) on the text extracted from the resumes like plots of experience and college ratings.
• Performed clustering of resumes based on the skill set of the people using a basic algorithm and by using LDA for topic modelling and giving distributions for different job profiles based on the skills.
• Finally, produce a user interface in flask with elasticsearch query support which gives you a decent search over all the fields of a resume. Like people from a certain college, people with a certain skill etc.
• Finally combined H2O and MLFlow to the working models and code and created an end-to-end UI implementation with Flask, Elasticsearch and Kibana. Also worked on modularizing the code.
• Researched libraries and created documentation to help the squad and add some value to their work. This product is supposed to help the recruitment team and will reduce the recruitment time

Feb 2019 – Feb 2020 · 1 yr · IIT Kanpur

• DAG aims to create a quality place where recruiters can find many talented, passionate, and dedicated students with a combination of data analytics skills and in-depth business knowledge.
• In addition to providing a forum for knowledge sharing, the group acts as a platform for students to interact with faculty and companies, and offers training in various programming languages and big data platforms.
• The group seeks to enhance students’ understanding of the role analytics plays in business leaders’ enterprises, finance, and other fields as well.

Dec 2018 – Jan 2019 · 1 mo · New Delhi Area, India

• Developed a independent platform for Social Media Analytics powered using Python and Django framework.
• Platform works for the public data from Youtube, Twitter, Tumblr and Reddit and private data from Facebook and Instagram.
• It shows you all type of insights of your data using the techniques of Sentimental and Time-series.
• This project is a successful implementation of natural language processing and database management.

Jun 2018 – Jul 2018 · 1 mo · IIT Kanpur

• Impact and Interaction analysis of SRI adoption on food security and well being of farmers using STATA
• Evaluation of Causal effects on treatment and training groups using TOT and ITT analysis with IV and OLS.
• Explored the untreated effect of SRI adoption on farmers in Bangladesh in the first-ever study of its type
• Results suggest that SRI adoption is strongly correlated with productivity, food security and well being

May 2018 – Jul 2018 · 2 mos · NYC Office, IIT Kanpur

• Investigated existence of universal voice prints that match a majority of users in text-independent speaker verification
• Customized 4 state-of-the-art implementations of Speaker Verification including Deep Speaker and Ivector
• Verified using evaluation metrics like Accuracy and ROC that Neural networks based models perform the best
• Tested all the models on Vox-Celeb dataset and ranked celebrities based on the uniqueness of their voice signature

Nov 2017 – Dec 2017 · 1 mo · work from home

• Created more than 100 video lectures for preparation of Joint Entrance Exam (JEE).
• Maintained good key performance Indicators like views on my videos, doubt removal etc.
• Facilitated learning for students using my knowledge and experience in teaching