Shashwat D.

I’m a DevOps/MLOps and backend-focused software engineer with a strong foundation in building scalable, cloud-native systems. With hands-on expertise in Docker, Kubernetes, GCP, Azure, and observability stacks like Prometheus, Grafana, and Loki, I specialize in designing resilient infrastructure and automating ML workflows to drive performance, efficiency, and reliability. At Donaldson, I led the end-to-end DevOps for ML inference workloads, setting up a production-grade Kubernetes cluster with observability baked in—handling over 500k logs/day across 10+ services. I migrated Azure Functions–based models into scalable, containerized APIs and architected the entire deployment pipeline for a Remaining Useful Life (RUL) prediction system. From Azure Event Hub to Snowflake via Snowpipe, I designed the data pipelines, implemented CI/CD, and introduced automated model monitoring and retraining—cutting issue resolution time by 40% and eliminating third-party dependencies. At Rapido, I drove automation at scale—provisioning GCP environments using Terraform and Ansible, streamlining CI/CD with self-hosted GitHub Actions, and implementing comprehensive quality gates with code checks and smoke testing. I contributed to the internal developer ecosystem by building tools like a static asset management service in Golang, enabling application-layer fault injection in Istio, and self-hosting Hoppscotch as a cost-effective API testing alternative. What sets me apart is my ability to blend backend development skills with infrastructure automation, enabling faster delivery cycles, reduced costs, and increased system resilience. Whether it's setting up ML infra, building reusable DevOps patterns, or enhancing developer productivity, I bring a proactive, problem-solving mindset with a passion for learning and scaling innovation. I'm excited by opportunities where I can build systems that not only run reliably but scale intelligently—bridging Dev and Ops with automation and ownership.