Available for senior cloud / DevSecOps / MLOps engagements

Karthick Raja Chinnathambi

Senior Cloud Architect · DevSecOps Specialist · MLOps Engineer

9+ years designing, hardening, and automating cloud-native platforms across AWS & GCP. Currently leading DevSecOps, MLOps, and the Tonik Autonomous Engineering Governance Platform at Tonik Digital Bank — the Philippines’ first digital-only bank. Researching agentic AI & geospatial ML for my M.Tech.

AWS Solutions Architect — Professional GCP Professional Cloud Architect Certified Kubernetes Administrator Terraform Associate M.Tech in Artificial Intelligence
01 — About

From network closets to autonomous engineering platforms.

I started out wiring up office networks in Coimbatore in 2014. Ten years later I’m architecting multi-cloud platforms that move money for a digital bank and training ML models that score customers in real time. The throughline has always been the same: infrastructure that’s secure by default and operates itself.

At Tonik Bank I lead the DevSecOps program — SAST/DAST pipelines, PCI DSS compliance, and the internal Plumbers Dashboard: a DevSecOps visibility and release-governance platform I designed with Aditiya Kadambari that enforces “build once, promote everywhere” across our pipelines. On top of that we’re building TAEGP — a twelve-agent autonomous engineering governance platform (with Pranesh Raja Mohan and Yugander) that lets AI safely operate against production AWS & GCP.

For my M.Tech in Artificial Intelligence at SRM I’m researching CropSense AI — a national-scale crop-prediction and farmer advisory system that integrates government land patta records with satellite ML for SDG 2 Zero Hunger. A recent drone-engineering internship rounded out the edge-ML side.

02 — Capabilities

The stack I build on.

MLOps & Data

  • Kubeflow
  • MLflow
  • TensorFlow
  • Vertex AI
  • KubeRay
  • FastAPI
  • BigQuery
  • GCS

DevSecOps

  • Semgrep
  • Talisman
  • SonarQube
  • MegaLinter
  • DefectDojo
  • Burp Suite
  • OWASP ZAP
  • PCI DSS

Cloud Platforms

  • AWS EC2/EKS/ECS
  • AWS Lambda
  • RDS
  • Route53
  • GCP GKE
  • BigQuery
  • VPC
  • S3 / GCS

Containers & Orchestration

  • Docker
  • Kubernetes
  • EKS
  • GKE
  • Helm
  • ECS / Fargate
  • Istio
  • ArgoCD

Automation & IaC

  • Terraform
  • Jenkins
  • GitHub Actions
  • Ansible
  • CodePipeline
  • CodeBuild
  • StepFunctions
  • Bash

Observability

  • Datadog APM
  • Prometheus
  • Grafana
  • CloudWatch
  • OpenTelemetry
  • ELK
  • PagerDuty
  • SLO design

Databases

  • PostgreSQL
  • MySQL
  • Redis
  • DynamoDB
  • BigQuery
  • S3 (data lake)

Languages

  • Python
  • Bash
  • YAML
  • HCL (Terraform)
  • Node.js (read)
  • Go (read)
03 — Experience

Where I’ve done the work.

Solution Architect — DevOps, Cloud & DevSecOps

Tonik Digital Bank · Jan 2022 — Present
  • Designed and operates the bank’s MLOps platform on Kubeflow + GKE, covering training, evaluation, and inference.
  • Built a FastAPI inference layer on BigQuery for real-time customer scoring and analytics.
  • Automated ETL pipelines with AWS Lambda + Terraform for downstream credit and risk systems.
  • Owns the DevSecOps program — Semgrep, Talisman, SonarQube, MegaLinter, DefectDojo wired into every CI/CD pipeline.
  • Deployed private domain routing and site-to-site VPN between AWS EKS and GCP GKE.
  • Monitors 100+ microservices with Datadog APM; ensures PCI DSS compliance across all cloud resources.
  • 50% faster release cycles · 30% lower model-training time via KubeRay · automated security posture.

DevOps Team Lead

MyAbhyas Pvt. Ltd. · Feb 2020 — Dec 2021
  • Automated container task scheduling with Jenkins + Terraform for zero-downtime microservice rollouts.
  • Improved app performance with Redis caching and CDN integration.
  • Led Prometheus + Grafana rollout — achieved 99.9% uptime and cut deploy errors 35%.

Network Administrator

Aim Window Software Solutions · Mar 2014 — Nov 2019
  • Managed IT infrastructure and applications, optimizing performance and availability.
  • Designed backup and disaster recovery policies for business continuity.
04 — Selected Work

Projects worth talking about.

Tonik Bank · with Aditiya Kadambari

Plumbers Dashboard — DevSecOps Visibility & Release Governance

The foundation TAEGP is built on. A centralized DevSecOps dashboard that aggregates CI/CD signals across Feature → Dev → Integration → Release → Production into a single source of truth. Adds a Release Readiness Score per build, enforces “build once, promote everywhere” artifact governance, and surfaces vulnerability lifecycle from Semgrep, SonarQube, and DefectDojo. Five-phase rollout culminating in AI-assisted deployment decisions and self-healing pipelines.

  • CodePipeline / CodeBuild
  • Semgrep
  • SonarQube
  • DefectDojo
  • ECR · ECS
  • Prometheus + PushGateway
  • Grafana
  • React
M.Tech AI · SRM University · SDG 2

CropSense AI — Crop Prediction & Farmer Advisory

AI-driven crop prediction and farmer advisory system integrating government land patta records (DILRMP) with Sentinel-2 satellite imagery, historical APMC prices, and IMD weather data. Farmers declare crops via a Flutter app or USSD (*99#); a CNN-LSTM yield predictor, XGBoost + Prophet demand forecaster, and Neo4j knowledge-graph advisor surface district-level supply forecasts 90 days before harvest. Targets Tamil Nadu / Maharashtra / UP pilot scaling to 10M farmers.

  • PyTorch (CNN-LSTM)
  • XGBoost
  • Facebook Prophet
  • Neo4j (Knowledge Graph)
  • Sentinel-2 · Cartosat-3
  • FastAPI · PostgreSQL + PostGIS
  • Flutter · USSD gateway
  • ISRO Bhuvan GEE

Modelled impact: 62% fewer price-crash events · 50% less post-harvest waste · +32% marginal farmer income

2026 · Authorized Security Audit

Plumbers Dashboard Security Audit (v2)

Authorized end-to-end security review of plumbers.tonikbank.com — the internal DevOps control plane fronting AWS account 988984599015. Identified and live-verified critical findings (SSM SecureString leakage, stored XSS via docx → mammoth ingestion, open-redirect on SSO callback, audit-trail attribution collapse to agent@system.local) and delivered a sprint-based remediation plan.

  • Burp Suite
  • nuclei
  • ffuf · sqlmap
  • Node.js / Express 5 audit
  • React / Vite audit
  • AWS IAM / SSM review
  • Azure AD / SSO review
Platform · Tonik Bank

End-to-End MLOps Pipeline

Production Kubeflow workflow at Tonik covering training, evaluation, registry, and deployment. Scalable FastAPI inference services on GKE serve real-time customer scoring and risk predictions; KubeRay distributed training cut training time 30%.

  • Kubeflow
  • MLflow
  • FastAPI
  • GKE
  • KubeRay
  • BigQuery
Platform · Tonik Bank

DevSecOps Automation Framework

Reference CI/CD pipeline integrating Semgrep, Talisman, SonarQube, MegaLinter, and DefectDojo on every commit. SAST + DAST run automatically; findings flow into a DefectDojo triage queue. Cut manual security-review effort by ~70% and accelerated release cycles 50%.

  • Semgrep
  • Talisman
  • SonarQube
  • MegaLinter
  • DefectDojo
  • GitHub Actions
Data · Tonik Bank

ETL Automation with AWS Lambda

Event-driven extract-transform-load pipelines feeding downstream credit and risk systems. Lambda + Terraform deployment cut processing time 40% over the prior cron-based pipeline.

  • AWS Lambda
  • Terraform
  • RDS
  • S3
  • StepFunctions
Resilience

Disaster Recovery for Stateless Workloads

Automated restoration of RDS snapshots and EC2 fleets via Jenkins + Terraform. Provides seamless failover for stateless application tiers across environments — backbone of the 99.9% uptime SLO.

  • Jenkins
  • Terraform
  • AWS RDS
  • EC2
05 — Education

Formal training.

2025

M.Tech in Artificial Intelligence

SRM Institute of Science and Technology, Chennai

Roll No. PA2312049010025

Case study: CropSense AI — AI for SDG 2 Zero Hunger

Currently graduating
2012

B.Tech in Information Technology

Anna University, Coimbatore

06 — Certifications

Credentials & continuing education.

Amazon Web Services

AWS Solutions Architect — Professional

Multi-account, multi-region architectures; advanced networking, security, and cost optimization.

Google Cloud

Professional Cloud Architect

Designing scalable, highly available, and secure GCP solutions — GKE, BigQuery, IAM, networking.

CNCF / Linux Foundation

Certified Kubernetes Administrator (CKA)

Production cluster operations — networking, storage, security, troubleshooting at scale.

HashiCorp

Terraform Associate

Infrastructure as code across AWS & GCP — module design, state management, drift detection.

Corizo · Drona Aviation

Drone Engineering Internship

20 Jun 2025 — 20 Nov 2025 · Dice ID CRZ138529. MSME & NSDC recognized. Edge ML + flight systems.

07 — Contact

Let’s build something.

Open to senior cloud architect, DevSecOps, and MLOps engagements — full-time or advisory. Best reached by email.