Sajal Garg

Data Scientist

Gurgaon, Haryana, India7 yrs 10 mos experience

Most Likely To SwitchHighly Stable

Key Highlights

Expert in machine learning and NLP techniques.
Proven track record in developing recommendation systems.
Strong background in data analysis and model deployment.

Stackforce AI infers this person is a Data Scientist specializing in E-commerce and Legal Tech with strong machine learning capabilities.

Contact

Skills

Core Skills

Machine LearningData AnalysisNatural Language ProcessingSoftware Development

Other Skills

Amazon Web Services (AWS)Bayesian Personalized RankingBi-LSTMClick ModelCollaborative FilteringComputer ScienceCustomer-Focused ServiceData VisualizationDigital IllustrationGitKafkaKeyword MappingMatrix FactorizationMySQLOCR

About

A resourceful Data Scientist with extensive experience in machine learning, NLP, and deep learning, I focus on leveraging advanced algorithms to analyze and interpret complex data. I specialize in designing and deploying models for personalized recommendations, using techniques such as collaborative filtering, matrix factorization, and Bayesian Personalized Ranking. My expertise also extends to natural language processing, where I employ deep learning models and transformer architectures to enhance text analysis and understanding. Proficient in tools like Python, TensorFlow, and scikit-learn, I am dedicated to advancing data-driven solutions and pushing the boundaries of AI.

Experience

7 yrs 10 mos

Total Experience

2 yrs 2 mos

Average Tenure

4 yrs 2 mos

Current Experience

Statusneo

Data Science Consultant

Apr 2022 – Present · 4 yrs 2 mos · Gurugram, Haryana, India

JIO Platform
Personalized Recommendation System
Developed and deployed a personalized recommendation engine across
retail e-commerce and media, incorporating various widgets that significantly increased user engagement and growth.
Through ongoing fine-tuning and model benchmarking, there was a 33% increase in click-through rates (CTR) and a 9% improvement in average order value (AOV).
Merchant Segmentation
Implemented advanced merchant segmentation models to categorize mer-
chants based on purchase behavior, product preferences, platform interactions, etc. using techniques like RFM analysis, and Topic Multi-label Classifier.
This led to an 11% increase in average order value and a significant reduction in user churn rate.
Search-based Relevant Product re-Ranking
Developed and implemented algorithms for search-based product re-
ranking using a click model, enhancing the accuracy and efficiency of search results, and boosting user engagement.

Computer ScienceCustomer-Focused ServiceMachine LearningData Analysis

Xceedance

Data Scientist

Apr 2021 – Apr 2022 · 1 yr · Gurugram, Haryana, India

Information Extraction from Contractual Emails
Implemented NER-based sequence labeling approach using Bi-LSTM and
word2vec embeddings.
Achieved an initial accuracy of 70%.
Information Extraction from Structured Documents
Leveraging YOLOv5 light-weighted object detection model for data ex-
traction from form structure documents and PDFs.
Achieved more than 85% accuracy by extracting data from the bounding boxes using OCR and Tabula.
Document Classifier
Implemented a bot for document classification.
Used advanced machine learning and NLP techniques, including XGBoost and n-gram TF-IDF vectorizers.
Achieved accuracy of 95.6%.

Alight solutions

Junior Research Analyst

Mar 2020 – Mar 2021 · 1 yr · Gurugram, Haryana, India

Conduct quantitative/qualitative analysis of different types of legal document images and unstructured data to retrieve the information using advanced ML algorithms.
Responsible for designing and maintaining an application that manages, controls, and monitors the different type of bots in the Pipeline to improve operational efficiency and implemented Kafka for distributed streaming of messages and dataflow between the bots
Implemented several object detection algorithm to detect multiple documents in a single image eg: Faster-RCNN, YOLO.
As a part of a 3 member team, we have developed a named entity recognition extraction system through keyword mapping which locates and classifies named entities in text into predefined categories.
Successfully implemented a bot for document classification using various advanced machine learning and NLP techniques including XGboost, N-gram TF-IDF vertorizers, etc. with an accuracy of more than 95%

Wipro limited

Project Engineer

Jul 2018 – Mar 2021 · 2 yrs 8 mos · Pune Area, India

Being a part of the core automation team, we delivered scalable and efficient solutions to the client for better management of resources and nullifying the human errors involved in the business.
Improved the decision-making efficiency of executives by developing an interactive dashboard to visualize various application performance data.
Developed a portal for users within the organization to automate day-to-day manual requests to update user credentials.
Cut report creation time from days to minutes by engineering a proprietary reporting software to automate the entire process.