Vidit Agarwal
Development
Uttar Pradesh, India
Skills
Machine Learning (ML)
About
VIDIT AGARWAL's skills align with IT R&D Professionals (Information and Communication Technology). VIDIT also has skills associated with Database Specialists (Information and Communication Technology). VIDIT AGARWAL has 7 years of work experience, with 5 years of management experience, including a low-level position.
View more
Work Experience
Lead Full Stack Data Scientist and Machine Learning Consultant
Ecosystem Consulting
December 2018 - Present
- Remote, IN Ecosystem Consulting is an start-up Consulting which specializes in delivering solutions to client from different domain Technology Stack: Python,SAP HANA, Looker,Hadoop, Power BI, Pandas, NumPy, SAP HANA ML,SciKit-Learn, Lambda,Tableu, AWS Glue,RedShift,Deeplearning Re-inforcement and NLP,GCP,pyspark,Scala,Sagemaker,Titan,AWS cloud,GraphQl,Rest,Jupiter Notebook,BERT,OpenAI,Haystack,LLM's Data Visualization & Predictive Analytics • Steering rapid model creation in Python using Pandas, NumPy, SciKit-Learn & plot.ly for data visualization • Constituting NLP models for Sentiment Analysis & MapReduce modules for predictive analytics in Hadoop on GCP • Designing a real-time contextual behavioral personalization system via econometric & ML to predict customer behavior • Created linear regression model to arrive at a potential transaction price for all future transactions. • Developed over 50 reports and dashboards on Power B I resulting in a 30% increase in operational efficiency. • Collaborated with cross-functional teams to identify data sources and streamline data collection resulting in a 25% reduction in report development time. • Improved dashboard usability by implementing interactive visualizations resulting in a 20% increase in user engagement. Statistical Modeling & ML Algorithms • Placed various machine learning techniques to build dynamic pricing models and maximize profitability • Led the development of a performance assessment & pricing analysis platform via k-NN Algorithm • Formed multivariate regression-based attribution models using ad stock analysis from the digital marketing data • Generated segmentation models using K-means Clustering in order to discover new segments of users Key Achievements • Established the Data Science division from scratch by recruiting, on-boarding & training a team of 8 Data Analysts. • Designed and developed reports and dashboards using Power BI Platform. Gathered data from various sources including SAP and Excel. • Proposed a reduction in negotiation time from an average of 3 weeks to 8 days & a 25% jump in closure • Improved report development through continuous collaboration with both stakeholders and developer teams across different business verticals. • Consulted over 23 clients on how to develop and maintain their personal self-service dashboard, which resulted in an overall increase of 19% in report accuracy. • Successful in an overall loss reduction of 10% on monthly revenue by implementing the loss minimization techniques • Migrated data transformation processes on Hadoop to reduce the data processing time by 25% • Developed a customer segmentation algorithm using Python to boost sales leads & increase market share by 28% Data Analytics and Predictive Modelling • Assisted in developing the Prioritization & Propensity model via Predictive Analysis, logistic regression & other statistical techniques on patients data. • Processed structured/unstructured data & gained insights by leveraging predictive modelling, statistics, ML & data mining • Rendering assistance in developing & deploying models while conducting predictive data mining to identify insights • Applied Machine Learning algorithms for identifying the most significant features across multiple datasets including medicinal,medical equipments datasets. • Utilizing deep learning packages like Tensorflow, H2O, lighGBM. Theano, Spark MLlib, etc. to develop ML models • Applied Bayesian Model Averaging (BMA) to combine individual keyword-level models. • Conducting data analysis to aid in developing models by employing machine learning and artificial intelligence to detect issues Sentiment Analysis,Clustering,Customer Segmentation,and Generative AI based NLP tasks • Led a project to develop a conversational AI chatbot powered by OpenAI's GPT-3 language model. • Developing NLP systems as per the requirements and implementing the right algorithms & tools for NLP tasks • Deployed statistical and machine learning models using the gathered data to evaluate sentiment and POS tagging via NLP • Created language models from text data & developed an autonomous Chat-bot based system • Transforming natural language data into useful features via NLP techniques to feed classification algorithms • Processed terabytes of data via Machine Learning & Predictive Modelling to develop dynamic solutions. • Implemented a convolutional neural network (CNN) for image classification, achieving an accuracy of 95% on a dataset of 10,000 images. • Conducting data analysis to aid in developing models by employing machine learning and artificial intelligence to detect issues • Trained a natural language processing (NLP) model using BERT for sentiment analysis of customer reviews, improving sentiment accuracy by 20%. • Designed and implemented a text summarization system using OpenAI's transformer-based models. • Reduced manual content curation time by 40% and improved the accuracy of summaries. • Leveraged OpenAI's Document Understanding models to automate document classification and data extraction processes. Specialist - Data Analytics Division BigFoot Retail Solutions Pvt Ltd |Delhi, IN Logistics Aggregator With a presence in 200+ countries Servicability in delivering Shipments, Shiprocket has delivered cost-effective IT solution Technology Stack: Python, Pandas, NumPy, SciKit-Learn, Lambda,Jupiter Notebook, AWS cloud,SAP HANA ML,Sagemaker,SP HANA Segmentation & Clustering • Applied large scale & low latency machine learning for non-parametric models & high-dimensional data visualization • Created multivariate regression-based attribution models & segmentation models using K-means Clustering • Utilized high dimensional data sets from users/media agencies/3rd-party apps via PCA, LDA & Kernel Approximations • Designed ARIMA to forecast demand and Random Forest model to predict parts at risk of being out of stock • Assisted in configuring performance trackers, reports, and dashboards for all promotional & marketing activities Key Achievements • Developed an additive scoring model for QSM and a logistic regression model to yield a K-S statistic of 51.5 • Deployed SGD, Logistic Regression, Random Forest, SVM, etc. for classification models to boost avg. click rate by 34% • Assisted the senior business analysts in devising statistical models for forecasting quarterly sales figures. Data Analyst Netway India Pvt Ltd |Gurugram, IN India's First Doctor's App and Astrology Service App having huge customer base across Globe. Data Analytics & Model Development • Directed model development, validation, testing and implementation of analytical products and applications • Suggested tweaks and new features for improving engagement by 29% & user retention by 34%. • Forecasted the quarterly sales figures and revenues to devise new strategies for the company. • Configured performance trackers, reports, and dashboards for all promotional activities. ML Algorithms & Statistical Analysis • Stationed advanced text mining algorithms to identify search intent latent in individual keywords. • Tested and implemented decision trees, random forests and ensemble models using bagging and boosting • Utilized hypothesis-driven approach to analyze A/B testing results & offer recommendations based on the test performance • Employed Principle Component Analysis to analyze collinearity and reduce the dimensionality of the dataset Data Analyst Corwhite Solutions Pvt Ltd |Gurugram, IN Mar '18 Nov '18- Nov '16 Feb '18-
Cloud Taxation and E-Filings
April 2015 - October 2016
- Data Analytics & Model Development • Directed model development, validation, testing and implementation of analytical products and applications • Designed database cubes successfully and star schemes using ER/Studio • Monitoring key metrics & understanding root causes of changes in metrics to identify opportunities for improvement. • Generating a monthly report on user behavior for the seniors within a stipulated time frame while ensuring accuracy. ML Algorithms & Statistical Analysis • Stationed advanced text mining algorithms to identify search intent latent in individual keywords. • Tested and implemented decision trees, random forests and ensemble models using bagging and boosting. • Employed Principle Component Analysis to analyze collinearity and reduce the dimensionality of the dataset.
Education
BITS PILANI
Master's
January 2023 - February 2024
