Areg Hovumyan

I am a robotics and machine learning integration engineer with hands-on experience building ROS2 nodes, real-time perception pipelines, and full-stack AI applications using C/C++, Python, and JavaScript. I focus on optimizing real-time systems and deploying scalable solutions across edge devices and cloud environments.

In competition and industry projects I developed production-ready object detection pipelines on a Jetson Orin Nano, improved detection speed by 60% and accuracy by over 400%, refactored legacy ROS2 code removing 34% redundancy, and established reliable wireless communications for unmanned ground vehicles. I also built AI companion and interviewer platforms that reduced infrastructure costs by 90% and implemented safety and moderation guardrails to cut harmful outputs by 30%.

I bring a pragmatic engineering ethos, strong debugging and monitoring practices, and experience across embedded robotics, cloud services, and AI/ML toolchains. I aim to deliver maintainable, high-performance systems that bridge perception, control, and user-facing applications.