8000 Karthik-099 (Karthik) · GitHub
[go: up one dir, main page]
Skip to content
View Karthik-099's full-sized avatar
🌐
Learning
🌐
Learning
24 followers · 40 following

Achievements

Achievement: YOLOAchievement: QuickdrawAchievement: Pull Shark

Achievements

Achievement: YOLOAchievement: QuickdrawAchievement: Pull Shark

Highlights

Block or report Karthik-099

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Karthik-099/README.md
Karthik. DevOps, Platform and Site Reliability Engineer. I keep production boring: observable, cost-efficient and self-healing. Open to DevOps, SRE and Platform roles. Remote or onsite. Bangalore, IST.
Kubernetes in production on AKS and EKS; observability with Grafana LGTM, Datadog and OpenTelemetry; cutting cloud bills without cutting reliability; GPU clusters; IEEE Access (Q1) published author

Email LinkedIn Portfolio Google Scholar Medium

Impact in production: 40% faster detection across 6 products, 50% lower infrastructure cost, 98% GPU cluster availability, zero SOC 2 / ISO 27001 audit findings.

About me

I'm a DevOps engineer who likes boring production: clusters that heal themselves, alerts that mean something, and bills that go down.

  • Now: DevOps Engineer at InCruiter, keeping an AI interview platform up under heavy concurrent load
  • Before: sysadmin for the GPU cluster at the Robert Bosch Centre, IISc Bangalore
  • Compliance: owned SOC 2 + ISO 27001 posture across Azure and AWS, with zero audit findings
  • Building: TerraGuard, LLM-assisted security review for Terraform
  • Upstream: a Datadog Agent fix in review
  • 1st of 200 teams at Fusion Techathon 3.0 · B.E. in AI & ML

Ask me about: LGTM stacks, spot instances without the tears, Slurm, canary rollouts.

 Animated Argo Rollouts canary: traffic shifts 10%, 50%, 100% to the new version while error rate stays under the SLO, then the release is promoted.

How I ship

Enterprise GitOps delivery pipeline. CI: commit, Gitleaks secret scan, unit tests, SonarQube quality gate, Trivy dependency scan and SBOM, Terraform and Kyverno policy as code, Docker build. Release: Trivy image scan, cosign signing and push to Docker Hub, ECR or ACR, config repo bumped to the image digest, Argo CD pull-based sync, Kyverno admission, Argo Rollouts canary, production on AKS, EKS, GKE or DOKS. Verify: Prometheus, Loki and Tempo feed an SLO analysis that promotes the release or rolls it back automatically, pages PagerDuty and reverts the config repo.

Things I've built

Project Stack What it shows
PaySentinel Jenkins · SonarQube · Trivy · EKS · Splunk a fraud-scoring service whose pipeline rolls itself back when Splunk spots post-deploy anomalies
TerraGuard Python · Flask · GPT-4o Terraform security review with severity-ranked findings, CWE tags and concrete fixes
GitOps Kubernetes Platform Argo CD · Argo Rollouts · Prometheus canary traffic shifting with automated rollback, all driven from Git
Event-Driven Microservices on EKS Terraform · GitHub Actions · EKS event-driven services provisioned end to end as code, observable from day one
API Health Monitoring on EKS Kubernetes · Terraform · multi-AZ autoscaled health checks for 10,000 endpoints
OpenShift AI Workloads OpenShift · Ansible · RHEL AI inference on Red Hat AI Platform with autoscaling and rollbacks
Spacelift IaC Orchestration · Ansible VM Health Terraform · Ansible · Spacelift stack dependencies and secrets done right · scheduled HTML health reports by email

Upstream

DataDog/integrations-core #25680 (in review): the Redis check no longer crashes with ZeroDivisionError on empty keyspace databases.

Research

Paper Venue
A comprehensive review on scaling machine learning workflows using cloud technologies and DevOps IEEE Access (Q1) · 2025
AI-Driven DevOps for intelligent automation in continuous software delivery pipelines IEEE ICECMSN · 2026
Secure communication protocols for multimedia things in IoMT Elsevier book chapter · 2026
Scalable machine learning solutions for solar energy output forecasting and PV performance improvement IEEE ICECMSN · 2026
Inorganic chemical reaction predictor using Random Forest and SVM IEEE AIDE · 2025

Toolbelt

Cloud, containers & orchestration

Azure, AWS, GCP, Kubernetes, Docker, Linux, Red Hat, Ubuntu
AKS EKS OpenShift Podman Helm

Infrastructure as code & delivery

Terraform, Ansible, GitHub Actions, Jenkins, Git, Bash, Python
Argo CD Argo Rollouts GitOps CAST AI Kubecost

Observability & security

Prometheus, Grafana
Datadog OpenTelemetry Loki, Tempo, Mimir Splunk SonarQube Trivy

GPU, AI/ML & data

NVIDIA H200 AMD MI300X Slurm Langfuse MLflow DVC
PostgreSQL, MongoDB, Redis ClickHouse

Need production that stays up and costs less? Let's talk.

Pinned Loading

  1. Real-Time-Parking-Space-Recognition-with-MLOps-Integration Real-Time-Parking-Space-Recognition-with-MLOps-Integration Public

    AI-powered smart parking solution uses deep learning and IoT cameras to detect available parking spots in real-time. It employs computer vision models to monitor parking lots, identify vacant space…

    Jupyter Notebook 1

  2. Ansible-VM-Health-Automation Ansible-VM-Health-Automation Public

    An Ansible-driven platform that collects virtual machine health and resource usage data and delivers scheduled, well-structured HTML reports via email.

    Jinja

  3. Scalable-API-Health-Monitoring-System-on-Kubernetes Scalable-API-Health-Monitoring-System-on-Kubernetes Public

    Production-ready API health monitoring system

    HCL

  4. spacelift-iac-orchestration spacelift-iac-orchestration Public

    End-to-end infrastructure automation on AWS using Terraform and Ansible, orchestrated with Spacelift for dependency management, secure secret handling, and repeatable deployments.

    HCL

  5. GitOps-Based-Production-Kubernetes-Platform GitOps-Based-Production-Kubernetes-Platform Public

    A production-grade Kubernetes platform implementing GitOps principles with Argo CD, canary deployments using Argo Rollouts, and comprehensive monitoring with Prometheus and Grafana.

    Shell

  6. pint pint Public

    Forked from cloudflare/pint

    Prometheus rule linter/validator

    Go 1

0