I design, automate, and scale cloud infrastructure with security and reliability in mind

Problem

Real-time fare estimation requires scalable infrastructure, low-latency APIs, and reliable deployments — none of which are trivial in a production environment. Manual deployments and lack of automated testing led to frequent downtime.

DevOps Focus

• •Containerized the ML model and API using Docker for consistent environments
• •Built GitHub Actions CI/CD pipeline with automated linting, testing, and deployment
• •Implemented Redis caching layer reducing API response time by 60%
• •Configured environment separation (dev/staging/prod) with managed PostgreSQL

Key Learning

Discovered that caching strategy mattered more than model optimization for user-facing performance. CI failures caught 3 critical bugs before production, validating the investment in automated testing.

Problem

Processing large-scale YouTube data required serverless architecture that could handle variable workloads without over-provisioning. Traditional ETL pipelines were too rigid and expensive at scale.

Architecture & Automation

• •Built event-driven data pipeline using AWS Lambda triggered by S3 uploads
• •Automated data cataloging with AWS Glue crawlers for schema discovery
• •Enabled ad-hoc SQL queries on S3 data lake using Athena (no database provisioning)
• •Implemented cost optimization with S3 lifecycle policies and intelligent tiering

Outcome

Reduced data processing costs by 80% compared to traditional EC2-based pipeline. Serverless architecture scaled automatically from 100 to 10,000 videos per day with zero infrastructure changes.

Problem

Healthcare applications require strict compliance (HIPAA), secure secrets management, and role-based access control. Traditional monolithic deployments made security audits and updates risky.

Security & DevOps Implementation

• •Implemented Azure Key Vault for encrypted secrets management (no hardcoded credentials)
• •Configured Azure AD RBAC with least-privilege access for patients, doctors, and admins
• •Built Azure DevOps pipeline with automated security scanning and compliance checks
• •Deployed to Azure App Service with automatic SSL, DDoS protection, and geo-redundancy

Challenge & Solution

Initial implementation stored database passwords in config files. Refactored to use managed identities and Key Vault, eliminating credential exposure. Added automated rotation policies for enhanced security posture.