Machine Learning Engineer (MTS-1) | Search Ranking | Advertising | Monetization | MLOps | Machine Learning
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
As a Machine Learning Engineer (MTS-1) in eBay's Search Science (Advertising) team, I work in the development of scalable machine learning pipelines and real-time advertising systems that directly impact user experience and revenue. My expertise lies in optimizing monetization for large online businesses through technologies like Scala, Spark, Hadoop, PyTorch, Airflow, Kafka, and Spark Streaming to name a few.
University of California, Santa Cruz
Master of Science (M.S.), Computer Science
January 1, 2016 – January 1, 2017
Tezpur University
Bachelor of Technology (B.Tech.), Computer Science and Engineering
January 1, 2012 – January 1, 2016
eBay
Staff Machine Learning Engineer (MTS-1) - Ads Engineering
January 1, 2023 – Present
eBay
Applied Researcher 1, Search Science - Advertising
January 1, 2021 – January 1, 2023
Excellerator (eBay Technology Recent College Grads)
Member Of The Board Of Advisors
September 1, 2018 – November 1, 2020
San Jose, California
eBay
Machine Learning Engineer 2, Search Science - Ranking
March 1, 2018 – January 1, 2021
University of California, Santa Cruz
Student Researcher at Natural Language and Dialog Systems Lab
September 1, 2017 – February 1, 2018
VERISIGN
Graduate Research Intern
June 1, 2017 – September 1, 2017
Washington D.C. Metro Area
University of California, Santa Cruz
Teaching Assistant
September 1, 2016 – December 1, 2017
Capsilon
Intern, Knowledge Capture Engineering
May 1, 2015 – July 1, 2015
Pune, Maharashtra
Tezpur University
Undergraduate Researcher Associate
August 1, 2014 – June 1, 2015
Big Data Analytics Lab, Tezpur University
Learning Dialogue response strategies using Deep Learning
September 1, 2017 – Present
Working in the Natural Language and Dialog Systems lab with Prof. Marilyn Walker to learn appropriate dialogue response strategies from a limited set of strategies using reinforcement learning.
Personality and Stylistic Variation of Retrieved Utterances
March 1, 2017 – June 1, 2017
We are modelling the personality of a conversational agent along the lines of popular film and television characters. We are developing approaches for varying the responses of a conversational agent based on the personality of a conversational agent. For developing different personality models, we are learning the personality of some famous film and TV characters from the IMDB subtitles dataset.
Recommender System using Probabilistic Soft Logic
March 1, 2017 – Present
Developing a novel method for converting a non-relational data into appropriate relational data. Developing a recommendation system to recommend subreddits from the Reddit data using Probabilistic Soft Logic.
User rating prediction from Yelp dataset
September 1, 2016 – December 1, 2016
(Group project in CMPS242 with three other students) • Modelled a user's rating behavior using his past rating data, using regression combined with multiple techniques. Predicting the rating a user assigns to a business from the generated model. • Designed a recommender system with Collaborative filtering model based on Yelp’s Millions of reviews • Processed and imported yelp raw data and converted user reviews into one-hot vector with Stanford CoreNLP • Analyzed and predicted user likelihood with Cosine Similarity and Linear Regression
SpectralBoost : Boosting performance of imputation algorithms using Spectral Biclustering
January 1, 2016 – May 1, 2016
Developed a new framework for boosting the performance of missing data imputation algorithms using spectral biclustering based recursive partitioning. We boosted three popular existing imputation algorithms with remarkable improvement in error. The project is the work of my undergraduate thesis. The project is with the Big Data Analytics laboratory, and I am advised by Prof. D.K.Bhattacharyya.
Deep learning for classifying network intrusions
July 1, 2015 – December 1, 2015
We used deep learning, using the Theano framework to classify intrusions in the KDD-Cup 1999 intrusion detection dataset. We used random forest based feature selection to select top-k features for better performance of the algorithm. This is the work of my 7th semester project. The project is with the Big Data Analytics laboratory, and I am advised by Prof. D.K.Bhattacharyya.
Gender and Language Prediction of Indian names using extracted natural language features
December 1, 2014 – Present
Mentor: Prof. U. Sharma Extracted multiple textual features of Indian names like ending letter, number of vowels, length of word etc. Evaluated the effectiveness of the features using SVM and selected the best features. Predicted Gender of a name given the language and vice versa and achieved an 85 percent accuracy.
Domain Specific Sentiment Analysis of News Headlines
August 1, 2014 – May 1, 2015
Developed a corpus-based approach for domain specific lexicon generation. Analyzed the sentiment behavior of sentences with the presence of positive and negative words. Computed the final sentiment scores using a hybrid approach combining existing thesaurus based sentiment scores as well as lexicon computed values. The project is with the Natural Language Processing lab and I an mentored by Prof. U.Sharma. Published a paper for 'Domain specific sentiment analysis in business news headlines' in NCCCIP-2015.
Convolutional Neural Networks in Tensorflow
DeepLearning.AI
June 24, 2026 – Present
Deeplearning.ai Tensorflow Developer Certificate
DeepLearning.AI
June 24, 2026 – Present
Neural Networks and Deep Learning
DeepLearning.AI
June 24, 2026 – Present
Sequences, Time Series and Prediction
DeepLearning.AI
June 24, 2026 – Present
Natural Language Processing in TensorFlow
DeepLearning.AI
June 24, 2026 – Present
End-to-End Machine Learning with TensorFlow on GCP
Google Cloud
June 24, 2026 – Present
Introduction to Finance - by University of Michigan, Ann Arbor
Coursera
June 24, 2026 – Present
Cultural Fit Analysis
The candidate demonstrates a strong cultural fit for a research-oriented ML Engineer role, given their extensive academic research, publications, and practical application of advanced ML techniques in industry. Their diverse project portfolio, ranging from imputation algorithms to sentiment analysis and recommender systems, shows a broad interest and adaptability. The long tenure at eBay, progressing through different ML roles, indicates loyalty and growth within an organization. The advisory role further highlights a commitment to community and mentorship. The certifications from DeepLearning.AI and Google Cloud show a proactive approach to continuous learning and staying current with industry trends.
Soft Skills & Operational Fit
The candidate's involvement as a Member of the Board of Advisors for eBay's Recent College Graduates program suggests strong mentorship, leadership, and communication skills. Their teaching assistant roles also indicate an ability to explain complex topics and support others. The project descriptions are clear and well-structured, indicating good written communication. However, without psychometric test results, a full assessment of work attitude, stress handling, and team collaboration is not possible.