Skip to main content

Aditya Nitin Bhavsar

Data Scientist

Talent
NJMember since September 22, 2026
PythonSnowflakeFeature EngineeringData FederationScikit-learnEnsemble Models (Random Forest, Gradient Boosting)Bayesian OptimizationHierarchical ClusteringPredictive MaintenanceAWSHigh-Availability Relational Databases (OLTP)FastAPIZero Data Loss CompliancePower BISelf-Service Analytics
·

Hire this person

Send a job offer directly to this candidate

PythonSnowflakeFeature EngineeringData FederationScikit-learn

About

Data Scientist with 3+ years of experience in machine learning, data pipelines, APIs, databases, and reporting across healthcare and IT. Skilled in Python, Scikit-learn, Spark SQL, FastAPI, Snowflake, and Power BI, with experience handling 3TB of claims data. Built models that improved AUC-ROC from 0.78 to 0.89 and reduced emergency visits by 28%. Currently working on database operations, reporting automation, AI monitoring, and cloud data integration.

Experience

Assistant IT (Data Systems & Infrastructure) - The Pierre, A Taj Hotel - NY

(2026-02)

  • Managing high-availability relational databases (OLTP) and distributed transactional systems to support real-time enterprise operations and zero data loss compliance.
  • Designing and implementing robust data backup pipelines, enforcing strict data governance and security protocols across all enterprise applications.
  • Collaborating with corporate engineering teams and enterprise vendors to map out upcoming cross-functional infrastructure upgrades and data migrations.
  • Architecting automated workflows for system diagnostics, currently actively transitioning manual IT reporting processes into dynamic, data-driven dashboards.
  • Commencing the deployment of AI-driven monitoring tools to analyze network telemetry, to transition the IT department to a predictive maintenance model.
  • Leading the architectural evaluation and phased integration of hybrid-cloud data federation layers, enabling seamless querying across on-premise legacy systems and modern cloud data warehouses to support real-time analytics without data duplication.

Data Scientist - Medvora Healthcare - India

(2021-06 - 2023-12)

  • Architected a hierarchical clustering model on patient demographic and claims data using Scikit-learn, identifying 8 high-risk patient segments and enabling targeted interventions that reduced emergency visits by 28%.
  • Automated feature engineering workflows using Spark SQL on 3TB of historical insurance claims data, improving model training efficiency by 65% and reducing feature development time from 3 weeks to 4 days.
  • Designed a REST API using FastAPI to serve real-time predictive models to 50+ hospital management systems, handling 10K+ concurrent requests with average response time under 200ms.
  • Optimized hyperparameters of ensemble models (Random Forest, Gradient Boosting) using Bayesian optimization, improving AUC-ROC scores from 0.78 to 0.89 across 12 disease prediction use cases.
  • Led the integration of Snowflake data warehouse with Power BI dashboards for 5 clinical departments, enabling self-service analytics and reducing adhoc reporting requests by 55%.

Education

Masters - Data Science - Pace University, Seidenberg School of Computer Science and Information Systems (2025-12)

Bachelor of Business Administration - Business Administration - Dy Patil, New Mumbai (2023-05)

Skills

Data Scientist

Reviews

Related