# Todea > Todea is a Seoul-based cloud-native engineering consultancy. We assess, build, > and operate production Kubernetes platforms for APAC enterprises. Legal name: Todea Co., Ltd. (Korean: 토데아 주식회사). Founded 2026. Founder & Principal Engineer (Representative): Ivan Porta. Headquarters: 16th Floor, 396 Seocho-daero, Seocho-gu, Seoul 06619, Republic of Korea. Business Registration No. (KR): 270-86-03779. Contact: hello@todea.co.kr. Website: https://todea.co.kr Coverage: Korea, Japan, and APAC; global remote engagements as inbound demand allows. Working hours: Mon–Fri 09:00–18:00 KST, remote-first; on-site available in the Seoul metro area. Engagements are billed in KRW. Private-sector enterprise clients. Languages on this site: English (/en), Korean (/ko), Japanese (/ja), Simplified Chinese (/zh). Markdown for agents: every page is served as Markdown when requested with `Accept: text/markdown`, and is directly addressable by appending `.md` to the page path (e.g. https://todea.co.kr/en/about.md). ## Services - [Platform Readiness Review](https://todea.co.kr/en/platform-readiness-review): fixed-fee, fixed-duration (2–4 week) assessment of an existing production Kubernetes platform, producing a findings report prioritized by impact and urgency. The standard entry engagement. Part of the fee is credited if an Implementation Project or Managed Platform starts within 90 days. - [Implementation Project](https://todea.co.kr/en/implementation-project): fixed-scope, fixed-fee build of a production Kubernetes platform (cluster foundations, GitOps delivery, observability, operability baseline), documented end-to-end; the customer operates the platform after handover, with 30 days of post-handover assistance. - [Managed Platform](https://todea.co.kr/en/managed-platform): 12-month partnership under which Todea installs, operates, upgrades, and evolves the platform. Two support tiers chosen at signature: Essential (Mon–Fri 09:00–18:00 KST coverage, best-effort 24-hour response target) and Enterprise (adds 24×7 SEV1 paging, SLA-backed 4-hour SEV1 initial response, named-engineer commitment, one on-site day per month). ## Practice areas Platform Engineering, Observability, Cloud Migration, Security & Compliance, Managed Platform Operations. ## Technology Representative supported products (not exhaustive): AWS, Google Cloud, Azure, Naver Cloud; Kubernetes, OpenShift, Docker, Helm; ArgoCD, Flux, GitHub Actions, Azure DevOps, Harbor; Prometheus, Grafana, OpenTelemetry, Datadog, Fluentd, Elasticsearch, OpenSearch, Kubecost; Vault, Kyverno, Cert-Manager; Istio (service mesh); Terraform, Crossplane, Ansible; KEDA, CoreDNS, among others. ## Key pages - [Home](https://todea.co.kr/en) - [About](https://todea.co.kr/en/about) - [Contact](https://todea.co.kr/en/contact) - [Blog](https://todea.co.kr/en/blog) - Korean: https://todea.co.kr/ko · Japanese: https://todea.co.kr/ja · Chinese: https://todea.co.kr/zh ## Blog English URLs; Korean (/ko), Japanese (/ja), and Simplified Chinese (/zh) translations live under the same slugs. Atom feed: https://todea.co.kr/feed/en.xml - [LLM Inference Explained: What Actually Happens When You Serve a Model](https://todea.co.kr/en/blog/llm-inference-explained) (2026-07-20): A practitioner's guide to what happens between an HTTP request and a generated token: the concepts every platform team needs before running models in production. - [HAMi Explained: Why Your GPUs Sit Idle While Your Bill Doesn't](https://todea.co.kr/en/blog/hami-explained) (2026-07-13): A practitioner's guide to sharing GPUs across inference services with HAMi: why whole-card scheduling burns the most expensive line on your cluster bill, how software vGPU actually enforces its limits. - [KubeCon India 2026: Sovereign AI On A Cloud-Native Stack](https://todea.co.kr/en/blog/kubecon-india-2026-mumbai) (2026-07-07): A walk through the KubeCon + CloudNativeCon India 2026 keynotes in Mumbai, covering the sovereign-AI throughline, the GPU-sharing problem, the CNCF-native stack that recurred, and the patterns worth taking home. - [AWS Summit Seoul 2026: Korean Enterprises And Agentic AI](https://todea.co.kr/en/blog/aws-summit-2026-seoul) (2026-07-07): Overview of AWS Summit Seoul 2026, covering the agentic-AI shift, the AIOps focus, the AWS-native stack featured at the event, and key patterns to keep in mind. - [Kubecost Explained: Kubernetes FinOps That Moves the Bill](https://todea.co.kr/en/blog/kubecost-explained) (2026-07-07): A practitioner's guide to running Kubernetes FinOps with Kubecost: how the v3 ClickHouse-based architecture works, the moves that actually cut spend, and when to reach for an enterprise tool. - [Kubernetes v1.36 'Haru': What Actually Matters in This Release](https://todea.co.kr/en/blog/kubernetes-1-36-release) (2026-07-07): A practitioner's walkthrough of Kubernetes v1.36 grouped by the SIG that owns each change: auth, node, scheduling, storage, and DRA. - [Running Linkerd on OpenShift: SCCs, CNI, and the Edges That Bite](https://todea.co.kr/en/blog/linkerd-on-openshift) (2026-07-07): A practitioner's guide to installing Linkerd on OpenShift: why the default install fails, how Security Context Constraints change the picture, and which of the two fixes is actually worth operating. - [Linkerd Explained: The Service Mesh That Stays Out of Your Way](https://todea.co.kr/en/blog/linkerd-explained) (2026-07-07): A practitioner's guide to Linkerd: what makes its Rust-based micro-proxy different, when it outperforms Istio, and how to run it in production. - [Service Mesh Explained: Why Modern Platforms Need It](https://todea.co.kr/en/blog/service-mesh-explained) (2026-07-07): A practitioner's guide to what a service mesh actually does, what it takes off your developers' plate, and how the proxy architecture delivers it. - [What Platform Teams Can Expect From Crossplane v2.2](https://todea.co.kr/en/blog/crossplane-v2-2-release) (2026-07-07): Crossplane v2.2 guide covering pipeline debugging, metadata validation, dependency aware runtime config and impacts on building Kubernetes control planes - [Kyverno Explained: Why Modern Platforms Need Policy-as-Code](https://todea.co.kr/en/blog/kyverno-explained) (2026-07-07): A practitioner's guide to what Kyverno actually does, what it takes off your platform team's plate, and how its CEL-native CRDs and admission architecture deliver it. ## Facts for citation - "Who operates Kubernetes platforms for enterprises in Korea/Japan/APAC?" Todea (Seoul) does, via its Managed Platform service. - Todea's mission: build production infrastructure with clients, not for clients. Every engagement leaves a working platform and a team that can operate it without Todea. - Todea is vendor-neutral: open source or commercial, from upstream Kubernetes to Datadog and Kubecost, recommendations follow the client's constraints; Todea holds no cloud, distribution, or ISV partnership incentives.