Python Design

Bhoomi Thakkar

+91-993**-***** ******.****@*****.*** linkedin.com/in/thakkarbhoomi

Education

Bachelor of Technology in Computer Engineering Aug. 2014 - Jun. 2018 Veermata Jijabai Technological Institute (VJTI) - 8.94/10.00 (Top 10) Mumbai, India Higher Secondary Certificate (HSC) Aug. 2012 - Jun. 2014 Pace Junior Science College - 91.69% (Top 1%ile state-wide) Mumbai, India Relevant Coursework

Data Structures, Algorithmic Design, Object Oriented Programming, Software Design Patterns, Machine Learning (ML), Natural Language Processing (NLP), Advanced Statistics (at Indian School of Business) Experience

Quantitative Research Associate Sep. 2020 - Current Citigroup Global Markets Limited Mumbai, India

• Working on constraint optimization of the Action PMFs developed in Citi’s Reinforcement Learning framework

• Performed parametric calibration and performance analysis (in Python, KDB+/Q) for behind and ahead quantity tolerances to control price slippage, effective spread capture and market impact for strategies like VWAP, TWAP Research Associate Oct. 2018 - Jul. 2020

Indian School of Business Hyderabad, India

• Gender Divide in Innovation - Estimated the effects of gender composition on patent acceptance rate, number of days to grant, citations, etc., for 2.5 million novel patents, using statistical and causal inference (using Stata and R)

• Text-based Firm Peer Networks and Classification - Developed technology clusters (using K-Means) and patent differentiation measures for 6 million patent abstracts vectorized using Python Gensim’s Doc2Vec. Examined the effects of patent differentiation on performance metrics (Tobin’s Q, Mkt Value) for 35,000 firms from 1997 to 2017

• Climate Risks Analysis - Examined the impact of climate risks, extracted from 10-K filings using TF-IDF and Latent Dirichlet Allocation (in Python), on performance metrics (from WRDS Compustat) for SNP-100 firms Software Development Intern Dec. 2017 - Feb. 2018

Hullo AI Mumbai, India

• Worked on auto-classification of responses for Hullo’s ’telebot’ using Naive Bayes and TF-IDF

• Developed a REST API for a report generation feature to portrait user journey through multiple campaigns, developed using Behaviour Driven Design approach for testing and development in Golang Student Trainee May 2017 - Jul. 2017

Samsung Research Institute Bangalore, India

• Identified patterns of digital transaction failure to increase customer satisfaction for Samsung Pay App (SPay)

• Load, store and process SPay logs (HTTP Request-Response Messages) using Logstash, Elasticsearch, and Kibana Selected Undergraduate Projects

Automated Image Captioning using Deep Learning Python, Keras (Tensorflow), NLTK

• Generated captions for images using LSTMs for sequence prediction (using pre-trained word embedding) and image-based features (extracted using VGG-16)

Identification and Classification of faults in Ball Bearings Python, pandas, numpy, scipy, statsmodel, scikit-learn

• Worked on fault diagnosis in rolling element bearings with data generated by the NSF I/UCR Centre for Intelligent Maintenance Systems, using logistic regression classification, with an accuracy of 92.5% Extra

PR & Design Executive - Technovanza, VJTI, 2015 - 2016 Editor - Nirmaan, VJTI, 2014 - 2015

Student Mentor & Volunteer - Spark-A-Change Foundation, 2014 - 2016 Jamnabai Adolescent Mentor (Student Council Body) - Jamnabai Narsee School, 2010 - 2012 Gold Medal for outstanding performance in Undergraduate (8.94/10.00, 2018), Higher Secondary (91.69%, 2014) and Secondary (95.71%, 2012)

Contact this candidate