Bobby Donchev

• Integrated and extended the mender client (Go based) to support custom device management. • Architected IoT device logging with mTLS (Go based)

• Led the design and development of a new AI platform using Go for both the CLI tool and backend services. The platform was a success and is now used by multiple teams across the company. • Oversaw the seamless migration and onboarding of users to the new platform, while significantly reducing provisioning time from one week to just 10 minutes. • Achieved an 80% reduction in maintenance and support tickets, paving the way for future enhancements such as a self-service user portal.

• Guided early-stage infrastructure architecture, technology stack, and team structures, establishing a strong foundation for the company. • Architected state machines for the payment domain using Golang, AWS Step Functions, which significantly improved the platform's availability and scalability.

Overhauled the system resilience and scalability by identifying and rectifying key bottlenecks. • Implemented idempotency for the services by introducing a new Golang sidecar. • Significantly improved the platform's resilience and handling capacity through strategic use of istio.

Successfully migrated services from GiantSwarm Kubernetes to Azure, enabling more efficient service management. • Architected a robust Apache Kafka messaging system ensuring smooth inter-service communication. • Facilitated rapid platform scaling and swift onboarding of new teams, contributing to the seamless migration of ~130,000 devices.

• Improved system stability by identifying and fixing connectivity issues between services and RabbitMQ cluster. • Successfully migrated RabbitMQ cluster to a managed service on AWS, significantly reducing system outages. • Achieved considerable improvements in mean time to recovery and mean time between failures, resulting in a more resilient platform.

• Led the successful deployment process on AKS using GitOps pattern, thereby ensuring a seamless transfer to on-premise cluster in the future. • Implemented a robust monitoring and alerting system using Prometheus, Grafana, and Alertmanager, facilitating continuous platform health check. • Despite the cancellation of the project, the work done was reused in other projects, showing the value and quality of the engineering practices established.

• Pioneered technology stack and system architecture, laying a solid foundation for the mobile banking app. • Developed a working prototype of the payment system, achieving key initial project milestones. • Implemented a microservice architecture with event sourcing and CQRS, ensuring a scalable and resilient system design.

Statistical Signal Processing Information Theory Convex Optimization in Communications and Signal Processing