Distributed ML Backend & Cloud Infrastructure

Architected and deployed a hybrid infrastructure supporting AI-powered mobile and web applications across both self-hosted and cloud environments.

​

Key features:

• Designed and deployed infrastructure supporting multi-environment (local + cloud) workloads across AWS and on-prem systems  

• Built backend systems capable of handling real-time image processing and AI inference pipelines  

• Developed architecture supporting scalable object storage and dataset management (S3 + MinIO)  

• Reduced system latency by migrating from cloud-hosted model inference to on-device CoreML execution  

• Enabled parallel model execution and validation workflows, improving diagnostic reliability  

• Designed infrastructure capable of scaling from local development to production AWS deployment (RDS, EFS, SageMaker)