02/10/2026 updated
100 % available
Data Scientist | ML Engineer | Python, SQL, PySpark, PyTorch, Scikit-Learn, Pandas
India
Worldwide
About me
I’m a Data Scientist + Machine Learning Engineer specializing in production-ready AI systems, combining strong ML expertise with practical MLOps and data engineering skills. I focus on building scalable pipelines, deploying reliable models, and delivering solutions that create real business impact.
Apache AirflowData AnalysisArtificial Neural NetworksComputer VisionContinuous IntegrationInformation EngineeringExtract Transform Load (ETL)Health CarePython (Programming Language)Machine LearningPerformance TuningMicrosoft Power BITensorFlowReproducibilityResource Allocation
I specialize in Data Science, Machine Learning Engineering, MLOps, and large-scale data pipelines, with strong experience in NLP, deep learning, and distributed processing. My core stack includes Python, PyTorch, TensorFlow, Scikit-Learn, PySpark, SQL, Docker, Kubernetes, MLflow, Airflow, ClickHouse, and Git-based CI/CD. I work end-to-end across the ML lifecycle: data ingestion, feature engineering, model development, production deployment, monitoring, and performance optimization.
Industries I’ve worked in:
Web Security / CDN (deep learning for web-attack detection, high-throughput log processing)
Healthcare Analytics (statistical modelling, resource optimization, Power BI dashboards)
General ML/NLP & Data Engineering (ETL, graph-based ML, computer vision, text similarity)
Notable Achievements:
Built and productionized an NLP-based web-attack classifier that reduced false positives by 2.7%, improving security signal quality for enterprise traffic.
Engineered high-volume ETL pipelines (200M+ records/day) using PySpark + ClickHouse, enabling real-time analytics and model training workflows.
Delivered ML models (XGBoost, neural nets) for healthcare resource allocation that cut operational costs by 3.3% with transparent insights (using SHAP).
Implemented production-grade ML deployments using Docker, Kubernetes, and MLflow, ensuring reproducible, scalable model serving.
Experience
Professional experience0 - 5 years
Freelancer since2026
Languages
EnglishNative speaker
Project history
NLP-Based Web Attack Classification (Professional):
Technologies: Python, PyTorch, PySpark, MLflow, Docker, Kubernetes, ClickHouse
Description:
Developed and productionized an NLP-based deep-learning model to classify malicious web requests at scale. Built an end-to-end pipeline covering data extraction, preprocessing, model training, evaluation, and deployment with MLflow and Kubernetes. Designed automated retraining logic and integrated performance monitoring.
Impact:
Improved attack-signal precision and reduced false positives by 2.7%, supporting more reliable security operations.
High-Volume ETL Pipeline (Professional)
Description:
Designed and maintained a high-throughput ETL pipeline for an API Discovery system, processing 200M+ records/day. Built scalable PySpark jobs for path normalization. Automated workflows using Airflow for daily ingestion and transformations.
Impact:
Delivered a reliable data backbone that enabled accurate API identification and cleaner analytics for engineering and product teams.
Healthcare Analytics & Resource Optimization (Professional)
Technologies: Python, PySpark, SQL, TensorFlow, Excel, Power BI, Hive
Description:
Built end-to-end data processing workflows to handle 10M+ healthcare records, including cleaning, validation, transformation, and feature engineering. Conducted statistical analysis (ANOVA, hypothesis testing, pattern discovery) to extract actionable insights for operational teams. Developed ML models (XGBoost, neural networks) to support resource allocation, and integrated SHAP-based explainability for business transparency.
Impact:
Improved operational decision-making and contributed to a 3.3% reduction in resource-related costs through data-driven recommendations and interpretable model outputs.