Skip to main content

246 posts tagged with "Engineering"

Engineering insights and technical deep dives

View all tags

How SNCF Keeps 200+ Kubernetes Clusters in Sync With Zero Drift — and Why the Same Loop Works on Three Hetzner Boxes
·Dora Noda·8 min

How SNCF Keeps 200+ Kubernetes Clusters in Sync With Zero Drift — and Why the Same Loop Works on Three Hetzner Boxes

SNCF, France's national railway, updates 200+ Kubernetes clusters every month with zero drift using Cluster API. Here's exactly how that reconciliation loop works — and why the same mechanism runs unmodified on three self-hosted Hetzner boxes.

self-hosting
PaaS
infrastructure
engineering
Talos Linux Meets Cluster API: The Immutable-OS Alternative to Ubuntu+kubeadm
·Dora Noda·9 min

Talos Linux Meets Cluster API: The Immutable-OS Alternative to Ubuntu+kubeadm

Talos Linux strips the Kubernetes node OS down to a dozen binaries, no SSH, and a single mTLS API — and it plugs into Cluster API through its own bootstrap and control-plane providers, right alongside CAPH on Hetzner.

self-hosting
PaaS
infrastructure
engineering
Railpack Replaces Nixpacks: What Railway's BuildKit Rewrite Means for Git-Push Builds Everywhere
·Dora Noda·9 min

Railpack Replaces Nixpacks: What Railway's BuildKit Rewrite Means for Git-Push Builds Everywhere

Railway replaced Nixpacks with Railpack, a BuildKit-based rewrite promising 38-77% smaller images — here's the real comparison against Cloud Native Buildpacks and what self-hosted tools like Coolify and Dokploy should do next.

self-hosting
PaaS
developer tools
engineering
Coolify vs Dokploy vs Kamal in 2026: Idle CPU, Multi-Server, and Why a Cluster-API PaaS Is a Different Bet
·Dora Noda·9 min

Coolify vs Dokploy vs Kamal in 2026: Idle CPU, Multi-Server, and Why a Cluster-API PaaS Is a Different Bet

Coolify, Dokploy, and Kamal charge wildly different resource rent to run your self-hosted PaaS. Here's the real idle CPU/RAM cost of each architecture, and why a Cluster-API-based platform is a different bet entirely, not a fourth entry in the list.

self-hosting
PaaS
infrastructure
cost-optimization
+1
Heroku Fir: The Kubernetes-Native Successor Nobody Outside Enterprise Can Use
·Dora Noda·10 min

Heroku Fir: The Kubernetes-Native Successor Nobody Outside Enterprise Can Use

Heroku's Kubernetes-native runtime, Fir, has been enterprise-only since its April 2025 GA, with a four-figure monthly price floor before you deploy anything. Here's what that architecture actually costs to run yourself, on Hetzner, with Cluster API.

PaaS
self-hosting
infrastructure
cost-optimization
+1
The 15% Failure Rate Nobody Benchmarks: Engineering AI Agent Tool Calls That Survive Production
·Dora Noda·12 min

The 15% Failure Rate Nobody Benchmarks: Engineering AI Agent Tool Calls That Survive Production

Benchmarks like SWE-bench test whether a model can solve a task. They never test what happens when the 1,000th tool call times out mid-pipeline. Here's the retry taxonomy, circuit-breaker discipline, and observability stack that separate agents that ship from agents that corrupt state.

AI agents
infrastructure
developer tools
engineering
Glamsterdam Slips: Ethereum's MEV Reform Hits Engineering Reality as ePBS Runs Late
·Dora Noda·10 min

Glamsterdam Slips: Ethereum's MEV Reform Hits Engineering Reality as ePBS Runs Late

Ethereum core developers have quietly conceded that Glamsterdam's ePBS implementation is running late, pushing the May-June 2026 mainnet target into Q3-Q4 and reshaping the L2 and MEV roadmap.

Ethereum
MEV
Flashbots
infrastructure
+1
Aave V4 Goes Live on Ethereum — But Its Tightest Governance Vote Ever Reveals DeFi's Growing Pains
·Dora Noda·7 min

Aave V4 Goes Live on Ethereum — But Its Tightest Governance Vote Ever Reveals DeFi's Growing Pains

Aave V4 launched on Ethereum mainnet with a hub-and-spoke lending architecture, but its binding governance vote passed with only 60 percent approval as core contributor BGD Labs departed, exposing the deepening tension between DeFi protocol engineering and token-weighted governance.

DeFi
Ethereum
DAO
engineering
+4
Ethereum's Glamsterdam Hard Fork Explained: How Parallel Execution and ePBS Target 10,000 TPS
·Dora Noda·10 min

Ethereum's Glamsterdam Hard Fork Explained: How Parallel Execution and ePBS Target 10,000 TPS

Ethereum's Glamsterdam hard fork targets H1 2026 with parallel execution via Block-Level Access Lists (EIP-7928), enshrined proposer-builder separation (EIP-7732), and gas repricing that could push L1 throughput to 10,000 TPS while cutting fees by 79%. We break down the architecture, the competition, and the risks.

Ethereum
scalability
MEV
blockchain infrastructure
+7
Showing 226–234 of 246 posts
Prev26 / 28Next