Pipelines, platforms & 99.9% uptime.
I'm Jaimit Patel, a DevOps engineer with 3+ years building CI/CD, Terraform and Kubernetes platforms across AWS, Azure, GCP and bare-metal data centres.
Hi, I'm Jaimit. I make infrastructure boring.
Deploys nobody worries about. Failovers nobody notices. Alerts that fire before users do.
I started in 2023 as a junior DevOps engineer on AWS while finishing my B.Tech in Computer Science. Since then I've run Kubernetes across three clouds, cut deployment time by about 40% with standardised pipelines, and now I'm building an on-prem data centre from the racks up: firewall, VLANs, Proxmox, HA PostgreSQL and end-to-end mTLS.
What keeps me in this work is the moment a fragile, manual process becomes a button anyone on the team can press safely. I like systems that explain themselves: everything in code, everything monitored, and no secret ever committed to git.
How I work
- 01Automate the second time.
Doing a task once teaches me the steps. Doing it twice means it becomes a pipeline, a Terraform module or a script.
- 02Not monitored, not in production.
Every service ships with metrics, logs and an alert that someone will actually read.
- 03Secure by default.
mTLS between services, secrets from a vault, least-privilege RBAC. Never a password in a repo.
- 04Design for 3 a.m.
Failover, backups and runbooks are part of the feature, and tested before anyone needs them.
Commit to production, 40% faster.
Jenkins, GitHub Actions and Bitbucket pipelines with Trivy scans and Argo CD GitOps. Scroll to run one.
- 1Commit & lintgit · pre-commit
- 2Testpytest · jest
- 3Build imagedocker · nexus
- 4Scantrivy · secrets
- 5Deployhelm · argo cd
↓ Scroll to advance, or
A data centre, from the racks up.
Click any node. Then break the database and watch Patroni recover.
The tools I run in production.
Numbers from the job, not the brochure.
faster deployments after standardising pipelines
less provisioning effort with modular Terraform
lower mean time to detect with Prometheus alerting
better AWS resource utilisation
git log --career
-
- Designed an on-prem data centre end to end: server sizing, switches, OPNsense, VLANs, DNS and Proxmox HA with NFS.
- Built HA PostgreSQL with Patroni, etcd, PgBouncer and pgBackRest for failover, pooling and recovery.
- End-to-end mTLS and secrets with OpenBao PKI, KV and Agent, delivered to PostgreSQL, RabbitMQ and Valkey.
- Observability with Prometheus, Loki and Grafana; Wazuh SIEM, JumpServer and OpenVPN access.
- HA Nginx reverse proxies and an in-house Nexus for artifacts, npm caching and Docker images.
- RabbitMQ quorum queues and Valkey Sentinel; GCP workloads on GKE, Cloud SQL and Compute Engine.
-
- CI/CD with Jenkins, GitHub Actions and Bitbucket, cutting deployment time by about 40%.
- Modular Terraform: about 50% less provisioning effort and 20% better AWS utilisation.
- Kubernetes with HPA, RBAC, ingress and storage at 99.9%+ uptime, Argo CD GitOps rollouts.
- Prometheus, VictoriaMetrics, Grafana and custom exporters, cutting time to detect by about 35%.
- Azure (AKS, ACR, Key Vault, App Gateway) and AWS (EKS, RDS, ElastiCache); ClickHouse ingestion.
- Trivy scanning and secret management built into delivery workflows.
-
- Provisioned AWS (EC2, S3, ECR, RDS MySQL) with Terraform.
- Automated operations in Python with Boto3.
- CI/CD workflows with GitHub Actions and Jenkins.
- Account access, billing and governance across AWS.
-
- In progress: AWS Certified DevOps Engineer, Professional.
- In progress: Certified Kubernetes Administrator, CKA.
- Languages: English and Hindi, both fluent.
Built after hours.
Keystone
A desktop SSH and infra manager: encrypted credential vault, chained jump hosts, tunnels and SFTP in one keyboard-driven app.
DevOps Dashboard
A Node.js app shipped through GitHub Actions with Trivy scans and Docker, deployed to Render.
Pulse Monitor
An Actix-web and Tokio service that exposes Prometheus metrics, packaged in Docker.
Where I'm headed.
Platforms that developers never have to think about.
I want to build internal platforms where any engineer can ship to production safely, in minutes, without waiting on ops or reading a wiki first.
- ✓ shipped2023 – 2026
Pipelines, Terraform and Kubernetes across AWS, Azure and GCP, at 99.9% uptime.
- ● in progressNow
An on-prem data centre at work, Keystone after hours, and CKA plus AWS DevOps Pro.
- ○ nextNext role
Platform or SRE work: golden paths, self-service environments, SLOs that mean something.
- ★ goalLong term
Lead a platform team, and grow Keystone into open-source tooling other engineers rely on.