---
title: "Ivan Porta (@gtrekter) | Hashnode"
description: "Founder @ Todea. Read the latest articles by Ivan Porta on Hashnode."
url: "https://hashnode.com/@gtrekter"
---

@gtrekter

# Ivan Porta

@gtrekter · Seoul, South Korea · Joined May 2026

Founder @ Todea

### [KubeCon China 2026: Capacity Recovered From The Same Hardware](https://todea.hashnode.dev/kubecon-china-2026-capacity-recovered-from-the-same-hardware?utm_source=hashnode&utm_medium=feed)

A walk through the KubeCon + CloudNativeCon + OpenInfra Summit + PyTorch Conference China 2026 keynotes and breakouts in Shanghai, covering GPU slicing in production, prefill and decode orchestration, cross-cluster rebalancing, and the patterns worth taking home.

11 min read · [todea.hashnode.dev](https://todea.hashnode.dev/kubecon-china-2026-capacity-recovered-from-the-same-hardware?utm_source=hashnode&utm_medium=feed)

[Sep 13](https://hashnode.com/posts/kubecon-china-2026-capacity-recovered-from-the-same-hardware/6aa6ca3e654429978c835534)

### [From Ingress to Inference Gateway: Gateway API and the Inference Extension](https://todea.hashnode.dev/from-ingress-to-inference-gateway-gateway-api-and-the-inference-extension?utm_source=hashnode&utm_medium=feed)

How Kubernetes networking became model-aware: the Gateway API resource model, the Inference Extension's InferencePool, and the endpoint picker that routes on KV-cache and queue metrics instead of round-robin.

19 min read · [todea.hashnode.dev](https://todea.hashnode.dev/from-ingress-to-inference-gateway-gateway-api-and-the-inference-extension?utm_source=hashnode&utm_medium=feed)

[Aug 27](https://hashnode.com/posts/from-ingress-to-inference-gateway-gateway-api-and-the-inference-extension/6a8fd785ca3ff895ff867ab8)

### [LLM Inference Explained: What Actually Happens When You Serve a Model](https://todea.hashnode.dev/llm-inference-explained-what-actually-happens-when-you-serve-a-model?utm_source=hashnode&utm_medium=feed)

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.

20 min read · [todea.hashnode.dev](https://todea.hashnode.dev/llm-inference-explained-what-actually-happens-when-you-serve-a-model?utm_source=hashnode&utm_medium=feed)

[Jul 24](https://hashnode.com/posts/llm-inference-explained-what-actually-happens-when-you-serve-a-model/6a637dc7abe32d5c27564b1c)

### [KubeCon India 2026: Sovereign AI On A Cloud-Native Stack](https://todea.hashnode.dev/kubecon-india-2026-sovereign-ai-on-a-cloud-native-stack?utm_source=hashnode&utm_medium=feed)

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.

6 min read · [todea.hashnode.dev](https://todea.hashnode.dev/kubecon-india-2026-sovereign-ai-on-a-cloud-native-stack?utm_source=hashnode&utm_medium=feed)

[1](https://hashnode.com/posts/kubecon-india-2026-sovereign-ai-on-a-cloud-native-stack/6a56b319e6527e8e0de7a134) [Jul 14](https://hashnode.com/posts/kubecon-india-2026-sovereign-ai-on-a-cloud-native-stack/6a56b319e6527e8e0de7a134)

### [Linkerd Explained: The Service Mesh That Stays Out of Your Way](https://todea.hashnode.dev/linkerd-explained-the-service-mesh-that-stays-out-of-your-way?utm_source=hashnode&utm_medium=feed)

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.

6 min read · [todea.hashnode.dev](https://todea.hashnode.dev/linkerd-explained-the-service-mesh-that-stays-out-of-your-way?utm_source=hashnode&utm_medium=feed)

[Jul 14](https://hashnode.com/posts/linkerd-explained-the-service-mesh-that-stays-out-of-your-way/6a56b2766ef474854d3df344)