Shirsendu Koner

Data Scientist | Machine Learning & Data Analytics Specialist

Email | Phone | GitHub | LinkedIn

[email protected]

91-8670108723

Kolkata, IN

About

Highly accomplished and results-driven Data Scientist with 2+ years of experience leveraging advanced machine learning, data analytics, and statistical modeling to drive business growth and optimize operational efficiency. Proven expertise in developing and deploying scalable AI/ML solutions, performing in-depth data analysis, and leading A/B testing initiatives. Adept at optimizing data pipelines, refactoring code, and translating complex data insights into actionable strategies. Seeking to apply strong analytical and technical skills to complex data challenges in a dynamic environment.

Work Experience

Data Scientist

Cars24 Services Pvt Ltd

Mar 2024 - Present

Driving auction performance and revenue growth through advanced data science methodologies and machine learning model deployment.

  • Enhanced auction performance by 8% through data-driven dynamic configurations and parameter tuning, leveraging simulation data and auction insights.
  • Designed and executed A/B tests and simulation-based Root Cause Analyses (RCAs) to validate pricing strategies, supporting regional rollouts and maximizing company revenue.
  • Refactored legacy code modules to improve readability, modularity, and reliability, incorporating robust error handling and fallback mechanisms.
  • Optimized system and data pipelines, reducing latency by 30% across pricing and auction workflows through API streamlining and memory optimization.
  • Developed and deployed a CatBoost classification model for real-time price acceptance prediction, utilizing hybrid feature engineering and real-time data pipelines, deployed via Docker/KServe on Google Kubernetes Engine for scalable production.

Data Scientist Consultant

Freelancing

Sep 2023 - Feb 2024

Provided data science consulting services, focusing on data visualization, LLM optimization, and machine learning model development.

  • Improved client data accessibility and usability by redesigning data visualization techniques, integrating statistical and informational graphics.
  • Developed and documented structured code for comprehensive client data analysis.
  • Optimized LLM performance through iterative prompt engineering and modification to achieve highly optimized results.
  • Engineered AI applications, including a document reader, using OpenAI GPT 3.5 and vector databases for efficient similarity searches on client data.
  • Developed an electrical signal image classification model by fine-tuning pre-trained Hugging Face transformers and PyTorch modules.

Data Science Intern

MakeMyTrip

Jan 2023 - Jun 2023

Contributed to data analysis, pipeline automation, and machine learning model development to enhance operational efficiency.

  • Enhanced work efficiency by 10% through the development and optimization of SQL queries on the Redash Platform.
  • Automated data extraction from Athena queries to S3 using Python pipelines and Redash.
  • Improved modeling and hyperparameter tuning processes by 10% through automated test report generation using Tensorboard and MLflow, with results stored in Google Sheets.
  • Resolved and maintained 3 critical Airflow DAGs, ensuring continuous automation.
  • Contributed to Neural Network model development, improving team efficiency by 5% through collaborative efforts.

Data Science Intern

OnePercent

Jan 2022 - Feb 2022

Gained foundational experience in data scraping, inventory management system development, and comprehensive data analysis.

  • Constructed comprehensive datasets by scraping data from 7 ed-tech websites, including GFG and Programiz.
  • Developed a foundational Inventory Management System.
  • Performed extensive Exploratory Data Analysis (EDA) and visualization on diverse datasets, including US Car manufacturing, Olympic Data, and Breast Cancer, achieving 100% data comprehension.

Education

Technology

NATIONAL INSTITUTE OF TECHNOLOGY

8.31 CGPA

Aug 2019 - Jul 2023

Certificates

Google Data Analytics

Coursera

Dec 2023

Data Science with Python

SimpliLearn

Sep 2022

Amazon Machine Learning Summer School

Amazon

Aug 2022

Advanced Excel Course

Udemy

Jul 2022

Crash Course on Python

Google

Jun 2022

Projects

GitHub Repository Analysis Tool

Jan 2023 - Dec 2023

Developed and deployed a tool to scan GitHub repositories under a specified username and identify the most complex projects.

Book Recommendation System (Popularity-Based & Collaborative Filtering)

Jan 2023 - Dec 2023

Created a hybrid recommendation system for books, combining popularity-based and collaborative filtering approaches to suggest personalized product choices.

Customer Service Request Analysis (NYC 311 Calls)

Jan 2023 - Dec 2023

Analyzed New York City 311 service request data to identify patterns, visualize complaint types, and derive actionable conclusions through robust data wrangling techniques.

Skills

Data Analytics & Visualization

  • Data Visualization
  • Tableau
  • Power BI
  • Data Analytics
  • Data Science
  • Data Wrangling
  • SQL
  • Analytical Approach
  • Advanced G sheet
  • Excel
  • SQL Joins
  • Common Table Expressions (CTEs)
  • Seaborn
  • Matplotlib
  • JSON Data Manipulation
  • AWS SageMaker

Machine Learning & Deep Learning

  • Machine Learning
  • Deep Learning
  • Neural Networks
  • TensorFlow
  • PyTorch
  • Keras
  • Fine-tuning Pre-trained Models
  • OpenCV
  • TorchVision
  • Pillow
  • Version Control
  • Model Deployment
  • Vision Transformer (ViT)
  • GPT-4
  • Diffusion Models
  • Prompt Engineering
  • Prompt Design

Deployment

  • AWS
  • Google Cloud
  • KServe
  • Kubernetes
  • Single-server Deployments
  • AWS Lambda
  • EC2 Instances
  • Docker

Programming & Tools

  • Python
  • PySpark
  • Object-Oriented Programming (OOP)
  • Multi-processing
  • HuggingFace
  • JIRA

Professional Skills

  • Problem Solving
  • Adaptability
  • Fast-paced Work Environment
  • Technical Communication
  • Teamwork
  • Customer Focus
  • Process Improvement
  • Actionable Insights
  • Curious Thinking
  • Passion

Documentation

  • Documentation Review
  • English
  • Written Communication
  • Presentation
  • Google Slides
  • PPT