I'm Tanat
Lokejaroenlarb.
I build and operate large-scale Kubernetes infrastructure for e-commerce marketplaces across Europe. On the Runtime team at Adevinta, I focus on platform reliability, developer experience, and cloud-native tooling. I write about real-world SRE incidents and platform engineering on Medium, and speak at conferences about what running Kubernetes at scale actually looks like.
30+
K8s Clusters
100K+
Pods
300K
RPS at peak
300+
Incidents led
Get to know me
About Me
Platform engineer with deep experience designing and operating large-scale Kubernetes infrastructure. Currently on the Runtime team at Adevinta, building SCHIP — an internal PaaS serving 1000+ developers across 30+ production clusters in 4 AWS regions. I am an AWS Community Builder in the Cloud Operations topic, a tech speaker, and a writer on real-world SRE incidents, platform patterns, and the intersection of LLMs and infrastructure.
Location
Barcelona, Spain 🇪🇸
Origin
Thailand 🇹🇭
Role
Staff SRE / Platform Engineer
Company
Adevinta
Community
AWS Community Builder — Cloud Operations
Education
BSc CS, KMITL — First-class Honours, Gold Medal
GitHub
github.com/InsomniaCoder
Where I've been
Work Experience
Mar 2024 — Present
Staff Site Reliability Engineer
Adevinta · Barcelona, Spain
- ▸Designed and evolved a cluster fleet management model, treating Kubernetes clusters as declarative, self-validating entities rather than snowflakes.
- ▸Built and operated custom Kubernetes operators acting as a control plane for cluster lifecycle, maintenance, and rollout orchestration.
- ▸Implemented progressive, SLO-gated cluster rollouts using reliability signals and blackbox tests to ensure safety before advancing changes across the fleet.
- ▸Established platform-level reliability standards including SLO/SLI definitions, alerting principles, and incident response expectations adopted across teams.
- ▸Designed instance management and reliability strategies to make maintenance automated, repeatable, and low-effort, significantly reducing operational risk.
- ▸Partnered with technical leadership to influence platform architecture, networking models, and operational strategy, balancing reliability, cost efficiency, and developer experience.
Mar 2023 — Mar 2024
Senior DevOps Engineer
Adevinta · Barcelona, Spain
- ▸Established and operationalised SLOs and SLIs for critical platform services, enabling objective reliability discussions and prioritisation.
- ▸Redesigned alerting and response workflows to shift from reactive paging to signal-driven operations.
- ▸Introduced self-healing mechanisms and automation to eliminate common failure modes and reduce operational noise.
- ▸Led the evolution of Kubernetes upgrade strategy, transitioning from Blue/Green to in-place EKS upgrades to simplify fleet operations.
- ▸Worked hands-on across Kubernetes, observability, and platform tooling to improve reliability, scalability, and cost efficiency at platform scale.
Oct 2021 — Mar 2023
DevOps Engineer
Adevinta · Barcelona, Spain
- ▸Operated and supported production Kubernetes clusters as part of a shared internal platform.
- ▸Defined reliability expectations (SLOs) for common platform services such as ingress, certificates, logging, and monitoring.
- ▸Built Kubernetes operators and automation to improve onboarding and reduce operational friction for platform users.
- ▸Integrated IAM, autoscaling, logging, and monitoring capabilities to accelerate teams from development to production.
Mar 2021 — Oct 2021
DevOps Engineer / Platform Infrastructure Owner
Monix · Bangkok, Thailand
- ▸Designed and operated AWS-based infrastructure including production-grade EKS, networking, and security foundations.
- ▸Built an internal developer platform covering CI/CD pipelines, identity management, secrets, and certificate automation.
- ▸Operated and maintained stateful production systems including Kafka and MongoDB.
- ▸Established cloud cost governance practices and introduced CNCF-aligned tooling such as GitOps, Kubernetes Operators, and Vault.
Jun 2016 — Nov 2020
Software / Platform Engineer
Exxonmobil · Thailand & Global
- ▸Led platform and DevOps enablement efforts for globally distributed engineering teams.
- ▸Designed Kubernetes-based platforms on Azure and OpenShift.
- ▸Defined Infrastructure-as-Code and CI/CD standards adopted across multiple teams.
- ▸Designed event-driven microservices architectures using Kafka and Redis.
- ▸Provided architecture guidance, reference implementations, and best practices to accelerate product delivery.
Education
Bachelor of Science (BSc), Computer Science
King Mongkut's Institute of Technology Ladkrabang (KMITL) · Bangkok, Thailand
2012 — 2015
What I work with
Skills & Technologies
Orchestration
Policy & Security
Observability
Cloud & Infra
Programming
AI / LLM
Writing
Blog Posts
Kubernetes · Incident
50K viewsWhen VerticalPodAutoscaler Goes Rogue: How an Autoscaler Took Down Our Cluster
Karpenter · Cost
Solving the Karpenter Price-Performance Trap with NodeOverlays
Observability · Platform
How we S(C)HIP Metrics for 1000+ Developers — Part 1
Observability · Platform
How we S(C)HIP Logs for 1000+ Developers — Part 2
LLM · Go · SRE
Developing My First SRE Helper LLM Agent Using LangchainGo
AI · Kubernetes
Running K8SGPT with Ollama Inside Your Kubernetes Cluster
Incident · SRE
Trial by Fire: Tales from the SRE Frontlines — Ep1: Challenge the Certificates
Speaking
Talks & Appearances
P99 CONF 2025
From Gatekeeper to Kyverno: Kubernetes Policy Management with Performance
2025
Sharing our journey migrating from OPA/Gatekeeper to Kyverno at scale — policy design, migration strategy, admission latency lessons, and how to avoid taking down your cluster with a webhook.
KubeFM Podcast
The Karpenter Effect: Redefining Kubernetes Operations
November 2025
How replacing EKS Managed Node Groups and Cluster Autoscaler with Karpenter transformed our Kubernetes operations, decoupled control/data plane upgrades, and saved €30,000/month.
KubeFM Podcast
Kubernetes Upgrades: Beyond the One-Click Update
May 2025
How Adevinta transitioned from blue-green to in-place Kubernetes upgrades for SCHIP, covering API deprecation tracking, cluster waves, PDB configuration, and ELB warm-up strategies.
DevOps Barcelona 2025
Observability to Resolution: The Journey Through a Production K8s Incident
2025
A live walkthrough of a real production Kubernetes incident — from alert to root cause — showing how observability tools (metrics, traces, logs) guide an SRE through diagnosis and resolution.
KyvernoCon 2025
Webhook Topology and Admission Latency: Lessons from Migration
2025
Real-world lessons from migrating Kubernetes admission webhooks at scale: how webhook topology affects cluster stability, how to measure admission latency, and how to survive the migration.
Say hello
Get in Touch
Interested in talking about platform engineering, SRE, Kubernetes, or cloud-native topics? I'm always happy to connect with fellow engineers — reach out via LinkedIn or Medium.