Work Experience

Spearheaded the development of a new doctors panel using Python and Vue, successfully migrating users from the old system without disruption.
Managed the end-to-end development lifecycle, from technical design to containerized deployment with Docker.
Ran the platform on AWS, operating managed EKS clusters and ECS for containerized workloads.
Architected and implemented Django database models, overcoming the project's most significant technical challenge to ensure data integrity and system functionality.
Built an AI-powered transcription pipeline using Deepgram, RAG, and Pydantic AI to transcribe doctor-patient sessions and generate structured SOAP notes.

Led the design and implementation of a company-wide cloud platform, progressing to lead engineer for platform architecture.
Engineered Kubernetes operators (CRDs) to automate ArgoCD authentication and authorization.
Led the design of a scalable microservice architecture for a high-volume delivery application, leveraging NATS to handle over 300k messages/sec.
Implemented KServe and Knative on Kubernetes to build ML pipelines, reducing resource consumption and boosting data gathering by 10%.
Deployed and managed Kafka on Kubernetes using the Strimzi operator; mentored teams on messaging best practices.

Optimized critical legacy applications (PHP) by migrating them to modern Go microservices, increasing availability by 20%.
Developed a real-time stream processing service to dynamically adjust pricing and balance demand/supply, resulting in a 5% increase in accepted rides.
Developed and maintained backend services in Go for real-time order processing and delivery tracking.

Secured and managed a ~$30,000 university research grant to establish the IoT laboratory and launch pilot projects.
Engineered a custom Reliable Layer-2 Transport Protocol (Stop-and-Wait ARQ with Retry logic) on top of raw NRF radio modules for guaranteed packet delivery.
Achieved stable performance for 1,000 concurrent IoT devices during load testing of the "Nahal" platform.
Built a scalable backend connecting 5000+ LoRa nodes for agricultural monitoring, utilizing Sentinel-2 satellite imagery for crop insights.

Developed a GraphQL API for the IoT Platform, enabling users to filter crop data with different filters.
Provided a data-processing pipeline using Apache Spark to ingest stock information, deployed on a serverless-like architecture using Docker Swarm.

Achieved an in-application video call solution using WebRTC based on the Pion Framework (Go); one of the earliest adoptions of Pion in the region, handling 1,000 concurrent calls via distributed design.
Increased application installation rate by 5% through the addition of stable video calling features.
Configured bare-metal infrastructure, including HPE ProLiant DL380 Gen9 servers with VMware ESXi.
Deployed services using Docker Swarm across staging and production environments.