Below is a showcase of key engineering projects that demonstrate my experience in Platform Engineering, Infrastructure Automation, and Agentic AI solutions.
1. Infrastructure AI Co-Pilot
| Domain | Agentic AI, LLMs, Operations Automation |
| Technologies |
LangGraph
RAG
Python
AWS API
GitHub Actions
|
Troubleshooting failing CI/CD pipelines and infrastructure alerts in a multi-region setup was highly manual, causing long developer blockages and higher mean time to resolution (MTTR).
Architected an autonomous AI Co-Pilot using LangGraph and Retrieval-Augmented Generation (RAG). The agent queries cloud logs (CloudWatch), cross-references codebase docs, and proposes (or auto-applies with approval) fixes for failing builds and infrastructure anomalies.
Developed the multi-agent graph architecture, build tools for log analysis, custom integration points with the AWS SDK, and the vector storage pipelines for developer runbooks.
- Reduced incident troubleshooting Mean Time to Resolution (MTTR) by 60%.
- Automated 75% of routine build error resolutions.
- Implemented secure, Human-in-the-Loop approval workflows for active infrastructure state modifications.
2. Self-Service Developer Platform Portal
| Domain | Platform Engineering, IaC, Cloud Security |
| Technologies |
AWS EKS
Terraform
Ansible
Docker
Python
|
Developer onboarding and new environment creation involved manual Jira tickets, taking between 3 to 5 business days per developer and leading to layout inconsistency across environments.
Designed a self-service developer portal integrating Terraform cloud templates and Kubernetes (AWS EKS) namespaces, allowing engineers to bootstrap isolated, compliant dev/test environments on-demand.
Built the core Python automation back-end, wrote reusable Terraform modules mapping company security compliance rules, and set up Ansible playbooks for automated server configurations.
- Decreased environment provisioning time from 4 days to under 15 minutes.
- Eliminated environment configuration drift across dev/prod environments by 100%.
- Implemented automated resource-expiry limits, saving over $45k/year in orphaned development servers.
3. Enterprise CI/CD Pipeline Standardization
| Domain | DevOps, Release Engineering, CI/CD |
| Technologies |
GitHub Actions
Docker
SonarQube
Prometheus
Grafana
|
Lack of uniformity across dozens of pipeline styles led to slow deployment times, security vulnerabilities, and no visual dashboard to track deployment health.
Migrated diverse legacy CI systems to a centralized, standardized repository of reusable GitHub Actions workflows containing auto-linting, SAST, and automated canary releases.
Designed reusable workflow templates, built pipeline health exporter metrics to Prometheus, and created real-time Grafana dashboards monitoring build times and deployment error rates.
- Reduced deployment cycle time by 70%.
- Achieved zero-downtime blue-green deployment strategies for customer-facing services.
- Added automated SAST scanning to 100% of active software projects.
Open Source & Community
AI Agent Operations Dashboard
- Role: Maintainer & Creator
- Description: A light-weight local monitoring utility that visualizes multi-agent graphs and token consumption metrics during runtime. Built with Python and Tailwind CSS.
- GitHub Repository
Custom Ansible Modules
- Description: Contributed enhancements to various community Ansible collections, improving compatibility with Windows platform management.
- Contributions Link
Interested in collaborating or building high-performance systems?
Let's Work Together