Help & Supportcareers@gravityer.com
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Machine Learning Engineer
Experienced Machine Learning Engineer and Scientist with 9+ years of experience designing, building, and shipping machine learning products at scale in the industry. I have led XFN projects in domain of display & search ads ranking, personalization, recommendations, and ranking algorithms. I have also worked on underlying infrastructure that powers these algorithms. As an academic, I have published in top-tier machine learning, data mining, and information retrieval venues including The Web Conference (formerly known as WWW), IJCAI, AAAI, CIKM, ECML-PKDD, and ICDM. Our work on investigating bias from spam and bots in social media won the best poster award at The Web Conference in 2016. Also, our work on learning node representations in dynamic attributed networks is one of the most cited papers of the CIKM 2017. My research interests are Machine Learning, Recommender Systems, Information Retrieval, Ads Ranking, Personalization, Search Ranking, Learning to Rank, Graph Mining, and Natural Language Processing and Understanding. Google Scholar: https://scholar.google.com/citations?user=hSeBK18AAAAJ&hl=en
Arizona State University
Master's Degree, Computer Science
N/A – Present
L.D. College of Engineering
Bachelor's Degree, Computer Engineering
N/A – Present
Sheth C.N. Vidhalaya
High School
N/A – Present
Apple
Staff Machine Learning Engineer/ Scientist
January 1, 2022 – Present
San Francisco Bay Area · Hybrid
Uber
Machine Learning Engineer
January 1, 2019 – January 1, 2019
San Francisco Bay Area
Scribd
Senior Machine Learning Engineer / Senior Applied Scientist
January 1, 2019 – January 1, 2022
San Francisco Bay Area
ipsy
Machine Learning Engineer / Scientist
January 1, 2017 – January 1, 2019
San Francisco Bay Area
Arizona State University
Graduate Research Assistant
January 1, 2015 – January 1, 2017
Tempe, Arizona
NCode Technologies, Inc.
Software Development Internship
January 1, 2013 – January 1, 2013
Greater Ahmedabad Area
L.D. College of Engineering
Undergraduate Research Assistant
January 1, 2012 – January 1, 2013
Greater Ahmedabad Area
Weighted Non-Negative Matrix Factorization and Ranking Game based Approach to Group Recommender Systems
October 1, 2015 – December 1, 2015
• We proposed a novel approach to group recommendation problem which used Weighted Non-Negative Matrix Factorization to predict the missing rating in the system and novel ranking game based aggregation strategy. • We show that proposed framework performs equally or better than state-of-the-art group recommender systems on evaluation metrics such as Group Precision @ k and Group Mean Average Precision @ k.
Learning to Detect Unseen Object Classes by Between-Class Attribute Transfer
April 1, 2015 – Present
• Developed a stacked ensemble classifier using direct attribute learning and multiclass SVM classifier aiming to analyze and improve learning when training data in particular is less. • Technologies Used: Python, scikit-learn, matplotlib, scipy, numpy
Predicting How Many Votes a Yelp Review will Receive
October 1, 2014 – December 1, 2014
• The problem is Kaggle Yelp recruiting competition. • Pre-processed the whole data using Python data analysis library (pandas) and extracted various features from such as no.of check-ins from check-in data. • Extracted various features such as Top 40 categories, Review Count, Average stars and the city from business data. • Pre-processed the review text, converted the text into lower case, removed the punctuation marks from review text using WordPunctTokenizer and Stemmed the review using PorterStemmer using nltk 3.0. • Identified various features such as length of the review, stemmed length of the review, ratio of original length to stemmed length, user-review star delta, business-review star delta. •Merged all features into the single feature vector and trained the feature vector using various regressors such as AdaBoost, RandomForest, Lasso, GradientBoosting and SupportVectorRegressor (SVR). •Achieved best RMSLE (Root Mean Square Logarithm Error) 0.46391 using Gradient Boosting Algorithm (Topper of the competition achieved 0.44321). •Technologies used: Python, scikit-learn, nltk 3.0, pandas, NumPy, SciPy
Flickr Network Analysis
September 1, 2014 – October 1, 2014
• Developed a web-crawler which collects data of Flickr users in BFS manner. The result is data-set consisting of 1.9 million Flickr connections between users in the form of edge-list and also anonymized the edge-list. • Computed Following properties of the Graph: • Plotted both in-degree and out-degree distribution of the graph and observed that network follows power-law degree distribution. • Computed the diameter of the graph, the number of bridges in the graph and minimum spanning tree of the graph. •Ranked the users in the graph using PageRank, Eigen-Vector centrality and in-degree. Also found most similar users using Jaccard Similarity, regular equivalence and adamic-adar. •Generated various network models such as Preferential Attachment model, Random Graphs and Small world model and compared it with original graph. •Technologies Used: Python, snap.py, NetworkX, matplotlib, gnuplot and MATLAB
Scribble It!
September 1, 2013 – March 1, 2014
• Scribble It! now has more than 10,000 downloads and rating of 4.1+ by more than 180 distinct users. • Developed an android application to draw and scribble on canvas. • Implemented various functionalities such as to change stroke color, change the background color, save the image to PNG format on local disk, import image from local memory to canvas, clear the canvas, share the image to various social networking website etc. • Implemented a 'Bluetooth mode' that allows the canvas to be shared with another user, allowing seamless simultaneous strokes from both users in real time.
Cultural Fit Analysis
The candidate's project diversity, ranging from Kaggle competitions and network analysis to group recommender systems and object detection, showcases a broad interest and adaptability in machine learning domains. Their professional experience at major tech companies like Apple and Uber, coupled with a focus on impactful ML solutions (e.g., improving engagement, reducing churn), aligns well with a high-performance, innovation-driven culture. The target role of ML Engineer is a strong fit given their extensive experience in ML model development, deployment, and leadership.
Soft Skills & Operational Fit
The candidate demonstrates strong leadership and problem-solving skills through their roles at Apple and Scribd, leading ML innovation and cross-functional initiatives. Their experience in A/B testing and deploying models to production indicates a practical, results-oriented approach. The multi-agent consensus framework project at Apple highlights innovative thinking and complex system design capabilities. The candidate's academic background also suggests a strong research and analytical mindset.