May 2021 – Apr 2024 · 2 yrs 11 mos · Gurugram, Haryana, India

• Industrialization of Systems for Model Training, Management & Deployment: Continuous
• Integration Framework, Cloud Infra Orchestration, Central Error Logging using Airflow, Docker,
• Azure DevOps & MLflow.
• Implemented CICD pipelines for building & deploying Docker containers using Azure
• DevOps to boost release by speeding development, decrease risks of deployment,
• quicker integration of user feedback & create trust in code quality.
• Developed Airflow DAGs using Python & deployed Airflow Docker containers in VMs
• to enable parallel pipeline runs and optimize throughput, track progress as they
• execute, set run time configurations to achieve dynamism.
• Leveraged Mlflow library for model’s parameters & metrics inspection, and also as an
• artifact registry.

Feb 2020 – May 2021 · 1 yr 3 mos · Greater Bengaluru Area

• HP is pioneer in print and imaging technologies – INK/ Laser/ LargeFormat/3D printing. HP generates huge volume of sensor/ telemetry data from different products in market PCs, Printers etc. As part of HP R&D, developed a common data platform to ingest and process printer telemetry data from different clients (printers, mobile apps, desktop applications).
• Developed Kinesis Firehose based data ingestion platform to load streaming data from various clients into data lake & process in batch manner.
• Developed data models, designed tables for Redshift and implemented Databricks/ Spark based ETL pipelines to process enrichment and telemetry sources of data.
• Analysed the existing technology environment and architecture, to develop technical recommendations to improve application performance.
• Created Docker images and co-ordinated with DevOps team to deploy pods in Kubernetes cluster (EKS) using Azure Codeway and Terraform.

Mar 2019 – Feb 2020 · 11 mos · Bengaluru, Karnataka, India

• Developed data models and created a Data Platform for Chicago based online food ordering & delivery marketplace that connects diners with local restaurants. It offers a method of accessing data warehouse in a consistent, supportable, fast and scalable way.
• Optimized logical data model for Hive and created data pipelines while delivering high data quality solutions that are testable and adhere to SLAs.
• Interfaced with business customers, gathered requirements and delivered complete data and reporting solutions owning the design, development, and maintenance.
• Developed Python/ Pyspark based ETL jobs which leveraged AWS's data technologies to pull the data from Cassandra, Salesforce, MySQL and other third parties into the data warehouse on Apache Hive.
• Identified downstream implications of data load changes during new data source integrations (e.g., data quality, code failure, future challenges).

Jun 2016 – Feb 2019 · 2 yrs 8 mos · India

• Build Enterprise Content Management solution for Australia based Bank. It can handle huge amount of structured, semi structured and unstructured data.
• Designed, developed & tested solutions for integrating data from varied sources.
• Developed Hive and PySpark based applications for file transformations.
• Monitored & troubleshot production bugs by identifying the root cause and implementing code fixes.