Yogesha K

Veritable: Accelerates patient eligibility verification and claim status checks by providing instantaneous updates in a simple and intuitive UI.

• Architected and implemented a multi-tenant architecture using Kubernetes CRD and Operators (Java/Quarkus), streamlining tenant onboarding and deployments, and reducing deployment time by 40%.
• Designed and implemented a robust billing management system integrating Stripe and Lago (open source), enabling seamless subscription management and accurate billing, facilitating rapid tenant onboarding, leading to a 3% savings in revenue.
• Spearheaded the development of AI-powered features for summarizing benefits and providing claim status insights, saving crucial time for healthcare admins in making decisions.
• Automated batch processing of eligibility and claim status checks, implementing robust retry and failure handling driven by configuration, reducing API costs by 25%.

Practifly: Digital marketing and utilities software for healthcare practices.

• Led backend development for a digital healthcare marketing platform, utilizing FastAPI, PostgreSQL, Celery, RabbitMQ, S3, Cloudflare, and Kubernetes, ensuring scalability and high availability.
• Mentored and provided technical leadership to a team of three backend engineers, fostering a collaborative environment, and improving team productivity.
• Collaborated with product and marketing teams to translate business requirements into technical solutions, ensuring alignment with marketing objectives, and delivering features that directly contributed to increased patient acquisition.
• Developed and implemented data pipelines for analyzing marketing campaign performance, providing actionable insights to healthcare practices, and improving campaign ROI by 18%.
• Rolled out an internal Celery worker monitoring and alerting tool using Prometheus and Grafana.

Muspell Health Data Platform: A data integration and centralized archival solution to better manage data and unlock better insights.

• Transformed data from 20+ EHR source systems into a normalized, scalable data warehouse, capable of handling millions of rows.
• Improved existing Spark jobs by optimizing queries, UDFs, and fine-tuning cluster configuration, reducing DBUs in Azure, resulting in a 20% reduction in cluster usage costs.
• Responsible for implementing and running data conversion jobs from FHIR to HL7 for ingesting into cloud-based EHRs, and custom PDF generation using LaTeX.
• Implemented Spark UDFs in Java for custom data decryption and DICOM image parsing.
• Mentored junior team members in developing good design decisions, development practices, and TDD.

Jeeves: A Just-in-Time AI Training Platform for EHR Users.

• Implemented APIs to ingest media and process them using FFmpeg to create streamable media.
• Transcribe media, Google Speech to Text, and index documents using Elastic Search, resulting in a better search experience.