Sayak Chatterjee

Data Analyst & Deep Learning Engineer

Kolkata, West Bengal 2+ yrs exp 84 · Excellent

About

Deep Learning Engineer and Data Scientist with proven expertise in data analysis, predictive modeling, NLP, and Computer Vision. Delivered significant value in a recent role at PwC by transforming and validating complex financial ERP data using SQL, Python, and Alteryx. Eager to apply advanced data analysis and machine learning skills to drive data-driven strategies.

Skills & Expertise (72)

Deep Learning Advanced

8.5/10

Years Exp

Python Advanced

8.4/10

Years Exp

Keras Advanced

8.0/10

Years Exp

NumPy Advanced

8.0/10

Years Exp

Pandas Advanced

8.0/10

Years Exp

OpenCV Advanced

8.0/10

Years Exp

TensorFlow Advanced

8.0/10

Years Exp

SQL Advanced

7.5/10

Years Exp

Git Advanced

7.5/10

Years Exp

Data Visualization Advanced

7.5/10

Years Exp

Power BI Intermediate

7.0/10

Years Exp

Root Cause Analysis ETL Predictive Analytics GANs Statistical Modeling A/B testing Statistical Analysis Preprocessing Data Cleaning EDA Exploratory data analysis Financial data analysis Autoencoders AI Tools Data Warehousing Data Modeling Data Governance Data Quality Data Pipelines Natural Language Processing Time Series Analysis Sentiment Analysis Web Scraping Hadoop Jupyter Notebooks Tableau Hugging Face Apache Spark PySpark Model Deployment Alteryx MySql Excel PyTorch scikit-learn C++ C# Java Dart HTML kotlin Predictive Modeling NLP Computer Vision Model Development Transformers Feature Engineering Classification Gradient Boosting Ensemble Methods Clustering Dimensionality Reduction PCA Supervised Learning Unsupervised Learning Hyperparameter Tuning Cross-Validation Neural Networks Convolutional Neural Networks CNN RNN

Work Experience

Data Analyst

PwC

Oct 2024 - Mar 2025

Transformed, cleansed, and analyzed 500K+ ERP records from Oracle, NetSuite, and JDE using Python, MySQL, and Alteryx, improving data readiness for six international audit teams and reducing data‐preparation time by 30%. Developed Python & MySQL test scripts for automated data‐integrity validation, detecting and resolving 98% of anomalies pre‐load and eliminating downstream audit discrepancies. Engineered ETL workflows in Alteryx to ingest mixed‐format financial inputs (Excel, CSV, PDF, fixed‐length), resolving data‐shifting errors and increasing pipeline reliability to 99.5% before performing transformations and cleansing using MySQL Server, Excel, Alteryx and UltraEdit on data sourced primarily from NetSuite, Oracle, and JDE. Collaborated with international audit teams and client stakeholders to define data extraction specifications, solidify data understanding, report results and clarify complex business rules, ensuring project objectives including timelines and deliverables were aligned with client needs from the onset. Documented data processing workflows rigorously from ingestion to final analysis reporting. Presented technical proposals and weekly progress updates to senior leadership and global clients, earning formal recognition for clarity and on‐time delivery. Led technical presentations to cross-functional teams, translating complex concepts into actionable insights. Collaborated with senior engineers to overcome critical technical challenges, contributing to reduction in project delivery time. Implemented Python scripts to automate routine data cleansing and validation tasks, reducing manual processing time by over 25% and minimizing the risk of human error. Created detailed summary reports and analytical dashboards in Excel and Power BI to visualize financial trends, key financial KPIs and discrepancies, enabling audit teams to quickly identify areas for further investigation. Instituted Git‐based version control and code review standards, reducing merge conflicts by 75% and accelerating team development cadence.

Deep Learning Engineer & Android Developer Intern

Resolute AI

Mar 2024 - Jun 2024

Assisted the development of cutting-edge deep learning models using TensorFlow and PyTorch, revolutionizing the company's approach to AI-driven solutions. Architected and deployed an innovative fashion recommendation system, leveraging advanced image processing techniques to boost user engagement. Pioneered the implementation of state-of-the-art YOLOv5 and YOLOv8 models, achieving 95% accuracy in custom object detection and counting in high-resolution video streams. Engineered a CNN for fabric pattern recognition that automated a manual process, resulting in a 30% increase in production efficiency. Designed and integrated a dynamic image processing pipeline using OpenCV and Streamlit, enabling real-time user interaction and enhancing product functionality. Optimized deep learning model performance through rigorous experimentation and hyperparameter tuning, consistently surpassing industry benchmarks.

Mobile Application Developer

Matainja Technology

Jun 2021 - Oct 2022

Implemented the application UI and developed reusable custom widgets to speed up application development. Integrated third-party libraries to interact with the back-end API. Integrated animations for the application UI and widgets. Designed dynamic and complex functionality and developed the application using the BLOC design pattern. Updated, monitored and maintained Git repositories and required application databases. Assisted the development of cutting-edge deep learning models using TensorFlow and PyTorch, revolutionizing the company's approach to AI-driven solutions. Architected and deployed an innovative fashion recommendation system, leveraging advanced image processing techniques to boost user engagement. Pioneered the implementation of state-of-the-art YOLOv5 and YOLOv8 models, achieving 95% accuracy in custom object detection and counting in high-resolution video streams. Engineered a CNN for fabric pattern recognition that automated a manual process, resulting in a 30% increase in production efficiency. Designed and integrated a dynamic image processing pipeline using OpenCV and Streamlit, enabling real-time user interaction and enhancing product functionality. Optimized deep learning model performance through rigorous experimentation and hyperparameter tuning, consistently surpassing industry benchmarks.