Internship Experience
-
Completed intensive hands-on training across AWS cloud services,
Docker containerization,
Kubernetes orchestration,
Terraform automation,
Linux administration,
CI/CD pipelines,
and monitoring systems.
-
Built automated deployment pipelines using Jenkins,
Docker, and GitHub Actions,
reducing manual deployment effort and improving release workflows.
-
Deployed containerized applications on AWS EC2 and Kubernetes clusters,
implementing scalable cloud-native infrastructure environments.
-
Practiced Infrastructure as Code using Terraform and Ansible,
enabling reproducible and version-controlled cloud provisioning.
-
Configured monitoring dashboards using Prometheus and Grafana
for infrastructure observability and performance analysis.
Projects
- Designed and implemented a fully automated CI/CD pipeline on AWS, reducing deployment time by ~70% compared to manual processes.
- Containerized the application using Docker and orchestrated automated build, test, and deploy stages via Jenkins and GitHub Actions webhooks.
- Deployed production-ready workloads on AWS EC2 behind Nginx reverse proxy with zero-downtime rolling deployments.
- Integrated environment-specific deployment configurations across Dev, Staging, and Production using parameterized pipeline stages.
- Architected a scalable microservices platform on Kubernetes, deploying multiple services with inter-service communication via ClusterIP and Ingress controllers.
- Implemented Horizontal Pod Autoscaling (HPA) policies, enabling the cluster to handle 3× traffic spikes without manual intervention.
- Built real-time monitoring dashboards in Grafana connected to Prometheus metrics — achieving full observability across all services with alerting configured for SLA thresholds.
- Managed application lifecycle using Helm charts, enabling repeatable, versioned deployments across environments.
- Automated provisioning of an entire AWS cloud environment using Terraform — including VPC, subnets, security groups, EC2 instances, and IAM roles — reducing infrastructure setup time by ~80%.
- Managed post-provisioning configuration and software installation using Ansible playbooks, enforcing idempotent, version-controlled infrastructure state.
- Implemented remote Terraform state storage in S3 with DynamoDB state locking, enabling team collaboration without configuration drift.
- Built a real-time AI system capable of detecting and classifying 7 human emotional states from live camera input using computer vision and ML models.
- Applied advanced preprocessing techniques (histogram equalization, Haar Cascade face detection) to improve detection accuracy by ~25% over baseline.
- Developed a multi-modal real-time recognition system combining hand gesture tracking (via MediaPipe) and facial emotion analysis in a unified camera-based interface.
- Optimized inference pipeline to achieve <50ms latency per frame, enabling smooth real-time interaction suitable for human-computer interaction use cases.