Hemanth D

About Hemanth D

I’m a Machine Learning / Data Scientist who loves turning messy data and promising models into reliable, observable, and scalable production systems. With an MS in Computer Science (UMass Boston, ’24) and hands-on experience at TCS, I’ve shipped data/ML pipelines and backend services across AWS, Azure, and GCP—using Python, PyTorch/TensorFlow, FastAPI/Flask, Spark/PySpark, Airflow, Docker, and Kubernetes. I care about CI/CD, metrics-first engineering (Prometheus/Grafana), and clean, well-tested code.Highlights:Built and maintained automated data pipelines on AWS (S3, Lambda, Redshift, Glue, Step Functions) with logging/validation and Jenkins-based CI/CD.Operated distributed/GPU environments (K8s, Docker, SSH) and automated workflows in a Linux-first stack.Delivered ML projects end-to-end—from data cleaning to model training/evaluation and serving—backed by monitors and alerts.Comfort across SQL/NoSQL (PostgreSQL, MySQL, MongoDB, Redis, DynamoDB) and streaming/big data tools (Kafka, Spark, Hive, Databricks).What I’m looking for:ML Engineer / MLOps / Data Engineer / Backend SWE roles where I can own pipelines and services from dev → prod, improve reliability/observability, and collaborate closely with data scientists and product teams.Let’s connect if you’re building ML platforms, data products, or high-throughput services in the NYC metro / remote.

Experience

1. Data Scientist

   Arthrex

   Mar 2024 — Present · Jersey City, NJ, US

   Applied ARIMA, SARIMA, Prophet, XGBoost, and LightGBM for time series forecasting of server demand, using Python,improving pipeline performance and demand prediction processes.Built automated ML pipelines using Apache Airflow, AWS SageMaker, and Git, integrating weekly model training, testing, anddeployment cycles.Developed interactive, geospatial dashboards in Tableau using GeoPandas, visualizing server load and demand patterns forClara AI and NRF initiatives.Performed segmentation and clustering analysis using Snowflake, Dask, and Pandas, identifying user behavior patterns forregional strategy alignment.Conducted A/B testing and hypothesis testing with Seaborn, Plotly, and statsmodels to evaluate feature changes andperformance metrics.Collaborated with DevOps teams to implement model monitoring and retraining triggers using custom Python scripts andintegrated them with AWS CloudWatch.Optimized data ingestion and processing using PySpark, standardizing schema parsing from Redshift logs to reduce data prepoverhead.Worked in Agile teams using JIRA and followed Scrum methodologies, aligning data science KPIs with business goals viacross- functional collaboration.

Education

• UMass Boston

  Master's degree, Computer Science

• REVA University

  Bachelor's degree, Computer Science