About Me

Let me tell you a few things...

Niroj

BIO

Results-oriented and highly skilled Software Engineer with 6+ years of background in creating and executing innovative software solutions to enhance business productivity. Highly experienced in all aspects of the software development lifecycle and end-to-end project management, from concept through to development and delivery. Consistently recognized as a hands-on and competent team member, skilled at coordinating cross-functional teams in a fast-paced, deadline-driven environment to steer timely project completion.

Software Engineer

ResMed 05/2021 – Present
Halifax, Nova Scotia​

Consent Domain Service

Technologies: Spring Boot, Java, Docker, Kubernetes, Helm, One Trust, Datadog, CI/CD, Open API Specification, Microservices, PagerDuty

  • Designed and developed a centralized service for managing and collecting user consent across multiple ResMed products using One Trust platform.
  • Built a flexible and scalable solution to support privacy policies and ensure compliance with global legal requirements related to data privacy.
  • Integrated the service with ResMed’s various platforms, ensuring a seamless process for obtaining consent for patient data, telehealth analytics, marketing compliance, and sleep health assessments.
  • Developed the application using Spring Boot, deployed through a CI/CD pipeline, and containerized it with Docker. Managed hosting and scalability using Kubernetes for secure, efficient deployment.
  • Implemented Datadog for monitoring and logging to ensure high availability and quick issue resolution in production environments.
  • Collaborated with cross-functional teams to design and implement solutions, improving efficiency in responding to legal and regulatory requirements.

    • Audit Log Service

    Technologies: Go, Python, AWS Lambda, Events, Streaming Data Pipeline, DynamoDB, CloudWatch

  • Developed and deployed an audit log service to track and log all user actions across ResMed's healthcare platform, ensuring full compliance with data privacy and security regulations
  • Built the system to support the capture of user activity logs, enabling transparent reporting and auditing of sensitive actions across products such as patient data access, health assessments, and telehealth services.
  • Integrated the service with existing cloud infrastructure (AWS) to collect and process logs, ensuring real-time reporting for security and compliance purposes.
  • Designed the audit log system with high availability and scalability in mind, using AWS DynamoDB for storage and AWS Lambda for serverless processing.
  • Enhanced system observability with CloudWatch and Datadog for proactive monitoring and quick resolution of issues in production environments.

    • Connected Health Platform (CHP)

    Technologies: TypeScript, Node.js, Bash, AWS (Amazon Web Services, SOLID Principles), Terraform, Serverless, Webpack, HL7 FHIR, Elastic Search, JWT Oauth 2.0, Docker, EKS, Kafka, Okta, PagerDuty, Angular, Hugo, Git, Postman, J-Meter

  • Spearheaded the design and development of the Connected Health Platform (CHP), a central solution for managing patient, provider, and device data across ResMed's suite of healthcare products.
  • Focused on enabling patient-centric workflows for devices like sleep, respiratory, and other medical devices while ensuring interoperability with external healthcare systems.
  • Implemented FHIR APIs as the standard for data exchange, ensuring consistent patient data access and enabling interoperability with EHR systems and third-party applications.
  • Designed and built the platform using AWS technologies including AWS FHIR Works, serverless architectures, AWS ECS Fargate, and SQS messaging to ensure scalability and reduced operational costs.
  • Ensured platform consistency with API versioning, backward compatibility, and integration with ResMed products like AirView, myAir, and Compliance Monitor.
  • Implemented feature-rich functionalities, including patient search, patient matching, and subscription notifications for real-time updates, improving healthcare workflow and reducing operational overhead.
  • Applied engineering and operational excellence principles, including unit, integration, and performance testing to ensure reliability and scalability via CICD pipelines.
  • Incorporated observability through standard logging, alerting, and instrumentation, enabling quick identification of issues and minimizing downtime.
  • Implemented OAuth2 OpenID Connect authentication with Okta and attribute-based access control for secure API authorization.
  • Developed SDKs and FHIR Implementation Guides to support internal and external integration with the CHP, promoting faster and easier product adoption across ResMed’s ecosystem.

    • Software Engineer

    GrowByData Services 07/2017 - 08/2019
    Kathmandu, Nepal​

    Technologies: Python, Pandas, NumPy, Django, SQL, Bash, E-Commerce, Rest API, MongoDB

  • Conceptualized and developed Data Exchange application that extracted client data provided through API and manipulated it into a format accepted by various online marketing platforms, increasing customer sales by 12% and improved reporting performance by 20%. Implemented Rule Engine to support different validation groups imposed by various online marketing platforms and monitored the export process.
  • Conducted in-depth analysis of customer requirements, leading to the development of efficient data migration tools. These tools streamlined the process of migrating customer data into various e-commerce platforms, resulting in reduced monthly costs for clients and an improved inventory management experience.
  • Spearheaded the development and maintenance of back-end applications and data management components. Implemented process improvements, leading to a remarkable 40% reduction in data loading time. Enhanced the ETL process, elevating data accuracy by an impressive 98%.
  • Designed and optimized MySQL databases, creating tables, stored procedures, and SQL functions to ensure optimal performance.
  • Transformed large SQL queries into stored procedures, significantly improving reporting performance and overall system efficiency.
  • Collaborated seamlessly with the QA team to address defects as part of the Sprint task, ensuring the delivery of high-quality software solutions.

    • Machine Learning Researcher

    Acadia Institute for Data Analytics 08/2020 - 08/2021
    Wolfville, Nova Scotia​

    Technologies: Python, Image Processing, Computer Vision, TensorFlow, Keras, Scikit-Learn, Linux, Bash

  • Spearheaded the collection of image and video data for training deep learning models, demonstrating a meticulous approach to dataset enrichment.
  • Conducted pre-processing of image datasets using advanced tools like OpenCV and skimage, ensuring data quality and relevance.
  • Implemented transfer learning with the MSCOCO dataset, fine-tuning the existing YOLOV3 model. This optimization resulted in a remarkable efficiency improvement of over 20%.
  • Orchestrated hyperparameter tuning during deep learning model training, leveraging tools such as Tensorboard for meticulous evaluation. Selected models based on optimal performance criteria.
  • Utilized an Intel RealSense depth camera to calculate the depth of each detected apple, contributing valuable insights into size determination.
  • Engineered and implemented a cutting-edge deep learning model for apple detection and tracking. This impactful solution significantly enhanced Scotian Gold Cooperative's decision-making processes related to production and marketing.
  • Achieved outstanding results with the final predictive model: an AP (average precision) score of 90.72 for apple detection, an MT (mostly tracked) score of 36.09 for tracking, and an F1 score of 94.97% for apple counting within the system. These metrics underscored the effectiveness of the developed models in real-world scenarios.