Open to opportunities

Chinmaya Meher

@chinmayameher

Python-focused Data Engineer building scalable ETL and data products, automating pipelines, and exploring ML/AI engineering.

India

Message

What I'm looking for

I’m looking to own end-to-end data pipelines and data products using PySpark and SAP Datasphere, while expanding into ML/AI system design—delivering reliable, automated workflows and measurable KPI improvements.

I’m a Python-focused Data Engineer with 3+ years of experience building scalable data systems, automation pipelines, and data products using PySpark and SAP Datasphere. I design reliable ETL workflows in layered architectures (Silver → Gold) and deliver curated datasets for analytics, while improving system performance and testing reliability through automation.

I also apply machine learning concepts through practical POCs and continuously expand toward ML engineering and AI system design. In Capgemini projects, I developed predictive cash collection using Logistic Regression and XGBoost, improving accuracy by ~28% via feature engineering and tuning, and I automate large-scale workflows by integrating Jira REST APIs and Xray with robust error handling—supporting onboarding and validation of 100+ data products.

Experience

Work history, roles, and key accomplishments

Current

Data Engineering Consultant

Current

Capgemini Technology Services Ltd

Aug 2022 - Present (3 years 11 months)

Built and optimized scalable PySpark ETL pipelines using a layered (Silver→Gold) architecture to deliver curated datasets for analytics. Developed SAP Datasphere data products with 10+ business KPIs and created a predictive cash-collection model (Logistic Regression, XGBoost) improving accuracy by ~28%.

Python SQL Pyspark ETL SAP Datasphere Data Modeling Xray API Integration Feature Engineering XGBoost

Postgraduate Researcher

NIT Rourkela

Jun 2021 - May 2022 (11 months)

Designed deep learning models (Autoencoder, Variational Autoencoder) for video anomaly detection on UCSDPed1 and UCSDPed2 datasets, achieving 84% accuracy and correctly identifying 11 of 12 anomaly types. Published 2 peer-reviewed research papers and built lightweight architectures suitable for real-time deployment.

Deep Learning Variational Auto Encoders Anomaly Detection Generative Models Model Evaluation

Education

Degrees, certifications, and relevant coursework

NIT Rourkela

M.Tech, Signal and Image Processing

2020 - 2022

Grade: CGPA: 8.80 / 10

M.Tech. in Signal and Image Processing (ECE) at NIT Rourkela, focusing on advanced signal and image analysis relevant to applied research and deep learning work.