Hariprasad Krishnamurthy

Designed and implemented end-to-end ML workflow of a recommendation and ranking system.
Designed and implemented experiment tracking tool across domain for various ml models.
Implemented large scale near real-time data pipelines for data warehouse Introduced & implemented Apache Airflow across domains within the organization.
Designed and Implemented various ingestion data pipelines from production databases such as PostgreSQL and DynamoDB.

Contributed to internal activities for overall process improvements, efficiencies and innovation.
Contributed to various infrastructure setup in GCP

Tech Stack - Python, Fast API, Airflow, Kafka, GCP, Dataflow, PubSub, Pyspark, AWS Glue, Kubernetes, Terraform.

Led Software development initiatives as Subject Matter Expert and primary point-of-contact for project management staff.
Improved system performance by making proactive adjustments and resolving bugs.

Health Record PDF generic parser

Worked on a generic parser for interpreting unstructured health reports generated by multiple diagnostic labs in several different formats. Reduced the manual report processing time by 80%.
Enabled Multi-tenancy in the platform

Introduced new design to enable multi-tenancy in the platform which reduced the infrastructure management effort and cost.
Data integration
Implemented syncing of large datasets from external partner sources using kafka stream api.

Technology Stack: Python, Django, celery, Kafka, Ruby on Rails, Rspec,

ReactJS, MySQL, Jenkins, Docker, Kubernetes and Microsoft Azure.

Telehealth Platform

Web-based telehealth solution with video conferencing, virtual waiting rooms and in-app chats.

Built backend as a serverless application on AWS lambda using Fast API and Pydantic.
Built Websocket support using AWS Websockets gateway on AWS Lambda. Integrated AWS Cognito as Authentication Service.

Automated test cases and deployment using drone and AWS

CloudFormation.
Technology Stack - Python, Fast API, Pydantic, Pytest, Postgres, AWS[Lambda, APIGateway, Websockets, Cognito, CloudFormation, Elasticache and RDS], Argo Workflow

Drove the re-architecture of Pharmeasy codebase from a monolithic architecture into microservices across different tech stacks - Python, Ruby and Java, with a strong focus on scalability and performance.
Built asynchronous communication between services using RabbitMQ as a message broker.

Devops tasks such as CI/CD pipelines in Jenkins and Ansible for Linux. Reduced infrastructure costs by 30%
Logistics Application - Designed and architected the initial in house logistics product, which was the first of the service oriented architecture from the existing monolithic application in Pharmeasy backend.

Retail and Warehouse Management System - Designed and developed inventory and warehouse product from requirement gathering to deployment following TDD standards.
Medicine Data Master [Product Catalog] - Developed product catalog which included medicine master along with over the counter products.

Technology Stack: Python, Django, celery, Pytest, Ruby on Rails, Rspec, ReactJS, MySQL, Jenkins, Docker, Kubernetes and AWS services.