Good to see you. I care about the whole path — how something is designed, shipped, and kept running in production — and I build in the open when I can.
I'm a DevOps and Platform Engineer with end-to-end ownership across the product lifecycle—how teams design, ship, deploy, operate, and improve software in production. My day-to-day work sits at the intersection of platform engineering, DevOps, cloud engineering, AI/ML deployments, and MLOps, with a focus on reliability, security, and measurable business outcomes.
What I do today: DevOps, platform engineering, cloud-native architecture, and production ML/MLOps infrastructure. What I've done: architecture design for distributed systems, and operating production platforms under real traffic and commercial load. What I'm building toward: full-stack application development, AI engineering, and stronger software engineering practices—I learn and write code in personal time through side projects and open source. I collaborate closely with product, engineering, and operations teams—on-site, hybrid, or remote, depending on what the role and team need.
Outcomes that matter to the business — not tool lists.
| Area | Result |
|---|---|
| Scale | Production systems for a US$160M+ annual business, 10M+ users, 400K+ DAU, and 4x traffic spikes during major sales |
| Reliability | 99.9% availability on mission-critical platforms; 76% MTTR reduction (30 min → 7 min) |
| Security | Zero Trust across production services, with SAST and DAST in the delivery path |
| Code delivery | 40%+ faster code delivery through CI/CD, IaC, and GitOps |
| Savings | US$120K+ cloud cost reduction (30% spend cut); 93% AdTech infrastructure cost reduction by replacing third-party with in-house |
| Revenue platforms | US$90M+ AdTech; US$5M+ POS across 200+ retail stores |
| Current | Working on platform infrastructure for an early-stage AI-Powered CI SaaS on GCP — SOC 2-enabled controls, GitHub PR-driven delivery (Atlantis + ArgoCD), up to 80% faster CI |
Working on platform infrastructure and deployment for an early-stage AI-Powered CI SaaS on Google Cloud—SOC 2-enabled controls, GitHub PR-driven delivery (Atlantis + ArgoCD), and production observability.
Impact: up to 80% faster CI · SOC 2-enabled Zero Trust · 13% spend reduction · every Kubernetes and infra change through a PR.
Owned production infrastructure and platform work for India's leading beauty e-commerce company—core commerce, AdTech, POS, data/ML platforms, and legacy admin systems. Hybrid GCP + AWS production (e.g. GKE workloads with Route 53, ALB, VPC, and WAF on AWS for AdTech and edge traffic).
Impact: US$160M+ business backbone · 10M+ users · 40%+ faster code delivery · 76% MTTR reduction · US$120K+ savings (30% cloud spend cut) · 99.9% availability through peak traffic.
Work I build in the open. This is the project I want you to look at first.
Open source, in active development. Grid is a tool you run yourself to manage infrastructure across clouds.
What should exist, what is actually running, and who changed it usually live in different consoles, scripts, and tickets. Grid is the place that holds that together. You describe the infrastructure you want. Grid creates it, changes it, and removes it the same way on every cloud, then shows you the result: what is deployed, what is in progress, and what needs attention.
You keep the infrastructure. Grid does not host it for you, and you can keep operating what it created if you stop using Grid.
The flow — how a change goes from what you asked for to something running — is on gridplatform.org. How it is built is in the write-up.
Achievement first — full docs →
| # | Project | Business outcome |
|---|---|---|
| 11 | Stealth Startup – Infrastructure & Compliance | Greenfield AI-Powered CI SaaS · up to 80% faster CI · SOC 2 + ISO 27001 |
| 1 | Purplle – E-Commerce Platform | US$160M+ revenue · 10M+ users · 99.9% uptime · 4x spike handling · US$120K+ savings |
| 2 | Purplle – Agentic RAG Sentiment | Faster brand decisions from influencer content · GPU + vector search at scale |
| 3 | Purplle – Data Science Infra | Marketing/ML platform · 50–60% manual-task cost savings |
| 4 | Purplle – Data Engineering Infra | 10M+ users' data to analytics · consumed by DS, Martech, SCM |
| 5 | Purplle – AdTech | US$90M+ revenue · 93% infrastructure cost reduction · in-house replacement for third-party |
| 6 | Purplle – POS (200+ stores) | US$5M+ revenue · 200+ stores · 99%+ uptime |
| 7 | Purplle – Observability | 76% MTTR reduction across 125+ microservices |
| 8 | Purplle – IaC Platform | 40%+ faster code delivery · standardized automation at scale |
| 9 | Purplle – Security & Compliance | Zero-trust across 125+ services · DPDP, ISO 27001, CIS, OWASP |
| 10 | Purplle – Legacy Admin Panels | Business ops without engineering dependency · US$160M+ panel backbone |
Production focus: Terraform, Ansible, Helm, Argo CD, Jenkins, GitLab CI, GitHub Actions · Kubernetes · OpenTelemetry
Cloud platforms (production experience on GCP and AWS):
| Area | GCP | AWS |
|---|---|---|
| Compute & containers | GKE, Cloud Run, Compute Engine | EKS, ECS/Fargate, EC2 |
| Networking & edge | Cloud Load Balancing, Cloud DNS, VPC, Cloud Armor | ALB, Route 53, VPC, WAF |
| Data & messaging | Cloud SQL, Pub/Sub, BigQuery | RDS, SNS/SQS, Athena |
| Identity & secrets | IAM, Secret Manager | IAM, Secrets Manager |
| Observability | Cloud Monitoring, OpenTelemetry | CloudWatch, OpenTelemetry |
Production hybrid: GKE workloads with AWS edge networking (Route 53, ALB, VPC, WAF) at Purplle; current AI-Powered CI SaaS platform work on GCP. Kubernetes and IaC patterns apply across both clouds.
Data & AI (deployments / infra): MySQL, MongoDB, Redis, Elasticsearch, Kafka, Qdrant, Vertex AI
Security: Zero Trust, RBAC, secrets management, compliance (SOC 2, ISO 27001, DPDP)
Learning & personal time: Python, JavaScript, TypeScript, React, Node.js, FastAPI—full-stack and AI application patterns through side projects and open source.
Personal-time learning and experiments—side projects and open source where I practice full-stack and AI application development.
| Project | What it is |
|---|---|
| Grid Platform | Open-source control plane · GitHub |
| TrendSignal | AI agent: YouTube trend analysis from a screenshot · MCP, FastAPI, GPT-4 Vision |
| Agility | React + MongoDB task management |
| SocialSplit | Real-time chat · Node.js, Socket.io, React |
| Portfolio / Blog / Bio starters | React templates I built and maintain |
- ✅ Google Cloud Associate Cloud Engineer
- 🎯 Preparing for Google Cloud Professional Cloud Architect
- 📖 Personal Blog — Engineering deep dives
- 🌐 Hashnode — Cloud, platform, and AI
- ✍️ Medium — Technical stories
Technical notes and deep dives—browse everything in DocHub:
| Repository | Content |
|---|---|
| Data Structures & Algorithms | DSA notes and solutions |
| DevOps Handbook | CI/CD, IaC, cloud-native, observability, security |
| Containerization Deep Dive | Docker, Kubernetes, container runtimes |
| Databases Deep Dive | Relational, NoSQL, analytical engines and managed services |
| Networks Deep Dive | Physical layers through cloud-native networking |
| System Design Concepts | Patterns, components, and trade-offs by use case |
| Commands and Cheatsheets | Essential commands and tool references |
| Data Engineering Deep Dive | Pipelines, warehousing, orchestration, platform engineering |
| Data Science & AI Deep Dive | Statistics, ML, deep learning, applied AI |
| Security Deep Dive | AppSec, IAM, cloud posture, forensics, defensive ops |
| Tooling & Frameworks Deep Dive | Frameworks, libraries, SDKs, platforms, toolchains |
Open to DevOps, platform, cloud, and MLOps roles — full-time or contract, on-site, hybrid, or remote (India and global).