OpenAI Deployment Layer: The Assistants API Precedent
OpenAI shipped the Assistants API in November 2023 and marked it for sunset 16 months later. That precedent is how to price the new deployment stack.
· 6 min read
Infrastructure
Hosting, edge compute, databases, and the pipes that hold modern apps together.
81 articles
701 Bad Internal Links Before 49 Good Ones
A phrase-matching script inserted 750 links across 269 MDX articles. Here is where they failed, and how claim-matching cut the review pile to 128.
· 7 min read
Cloudflare Web Analytics via GraphQL: the siteTag Filter
Why accountTag and siteTag differ in rumPageloadEventsAdaptiveGroups, how the limit argument truncates silently, and which fields we did not verify.
· 6 min read
Bing Webmaster API's 100-URL Cap: 289 URLs, Three Days
Quota is 100 a day against 1300 a month, GetQueryStats was still empty at the end, and two error shapes will kill a cron job.
· 6 min read
IndexNow Batch Mode: 286 URLs Per Publish Down to 0
Bing flags full-sitemap submissions as batch mode. The ~40-line lastmod diff that fixes it, the precondition it needs, and two ways it silently breaks.
· 6 min read
Astro middleware can't serve 410 under output: 'static'
We deleted 434 articles. Middleware runs at build time and 404s in production; two prerender:false routes fix it on Cloudflare Workers.
· 7 min read
Workers KV's 60-Second Consistency Window
It's a cache TTL you cannot lower, not a replication delay. It bites cached nulls, uniqueness checks, and multi-key updates. When to swap in a Durable Object.
· 6 min read
Driving Chrome With the DevTools Protocol, and When Not To
CDP gives a scheduled agent a real browser when a site ships no API. The memory and wall-clock cost, plus four failure modes that only surface on a cron.
· 7 min read
Cache Races Make IndexNow Miss Pages You Just Shipped
A deploy API returning 200 does not mean the new URLs are reachable at the edge. Here is the verification sequence that stops wasted submissions.
· 6 min read
Object Storage Lifecycle Policies: Cut Cost, Keep Data
Audit an S3-compatible bucket, then write lifecycle rules that avoid transition fees, minimum-duration charges, and versioning traps that raise bills.
· 7 min read
k3s vs Talos vs a Managed Control Plane in 2026
Three options that solve different halves of the same problem. What each costs you in money, upgrade work, and 2am debugging on a three-node cluster.
· 6 min read
Postgres Pooling 2026: PgBouncer vs Supavisor vs Drivers
Postgres forks a process per connection, so pooling is not optional past a few dozen clients. Plus what transaction mode breaks.
· 7 min read
Blue-Green Deployments Without a Platform Team
No Kubernetes and no service mesh. A working setup built from a reverse proxy, two ports, and a shell script.
· 7 min read
CDN Edge Caching for Application Developers
Not just static assets: with the right Cache-Control headers, a CDN can serve API responses, authenticated content, and dynamic pages.
· 7 min read
Database Backup Strategies That Pass a Disaster Drill
Backup scripts that create files can still fail to restore. Scheduled restores, WAL archiving, and the three things your backup must prove it can do.
· 8 min read
Infrastructure as Code for Solo Founders
You do not need a Terraform monorepo or a dedicated infra engineer. One main.tf, a state backend, and a GitHub Actions workflow on push is enough.
· 7 min read
When Serverless Becomes More Expensive Than a VPS
Above a certain traffic volume, per-request billing costs multiples of a $6 VPS. Here is the crossover math.
· 7 min read
Caddy vs Nginx in 2026: Automatic HTTPS
For solo developers and small teams: certificate management, performance trade-offs, config ergonomics, and when switching actually pays off.
· 7 min read
Hetzner vs OVH for Side Projects: Bare-Metal Value in 2026
Pricing models, bandwidth, and hardware compared, plus the trade-offs that matter for a solo developer.
· 7 min read
Bun vs Node.js in Production: What Actually Changes in 2026
Install speed, native TypeScript, and built-in tooling versus the compatibility and observability gaps that still bite in real deployments.
· 7 min read
Coolify vs Dokploy: Self-Hosted PaaS Compared for 2026
For solo devs running their own deployments: how Coolify and Dokploy differ in architecture, setup, and resource use, and which one to pick.
· 7 min read
Turso vs Neon: Serverless SQLite and Postgres in 2026
How the two compare on latency model, database branching, cost shape, and lock-in risk, so you can pick one for your workload.
· 7 min read
Kamal 2 review: containers without Kubernetes in 2026
37signals' Docker-over-SSH deploy tool: what kamal-proxy changed, where it fits, and where it hands the hard problems back to you.
· 6 min read
Caddy vs Traefik vs nginx Proxy Manager
We migrated three production stacks across Caddy 2.8, Traefik v3.1, and nginx Proxy Manager 2.11. Where each earns its keep, and where it bites.
· 7 min read
K3s vs MicroK8s vs k0s: lightweight Kubernetes in 2026
Compared on memory footprint, default add-ons, and HA story, plus which team shape each fits. Operational opinions, not synthetic benchmarks.
· 6 min read
Coolify vs Dokku vs CapRover: self-hosted PaaS in 2026
We deployed the same three-service app to a $12/mo Hetzner box: architecture, memory footprint, and ops cost of each.
· 7 min read
Block Hands Goose to the Linux Foundation
What vendor-neutral governance means for teams choosing between LangChain, AutoGen, and Goose - and the lock-in risk most overlook.
· 6 min read
Fly.io Edge Platform Review: Deploy Apps to 37 Regions
We deployed a Go API and Next.js app to measure cold starts and latency, comparing DX with Railway, Render, and Heroku, plus fly.toml and WireGuard deep-dive.
· 9 min read
OrbStack Deep Review: Replacing Docker Desktop on macOS
We moved 18 containers off Docker Desktop on an M1 Max MacBook Pro and measured memory, idle CPU, cold starts, and Docker API compatibility.
· 8 min read
Temporal: Durable Execution That Survives Process Death
We built payments, onboarding, and AI orchestration on Temporal, then compared it with Step Functions and job queues. Includes the SDK learning curve.
· 10 min read
Turso libSQL: The SQLite Fork With an Edge Replication SDK
Notes from running embedded replicas across 3 regions in a TypeScript pipeline, plus how it compares to Cloudflare D1 and PlanetScale.
· 9 min read
Upstash Review: Serverless Redis and Kafka Benchmarked
Latency measured from 3 regions vs AWS ElastiCache and Confluent Cloud, plus the Redis REST API, Kafka HTTP bridge, and where per-request pricing wins.
· 9 min read
NVIDIA CUTLASS: CUDA Templates for AI Linear Algebra
A close read of the header-only library behind much of modern AI infrastructure: its kernel hierarchy, where CuTe and the Python DSL fit, and when to use it.
· 7 min read
ROCm in 2026: Why PyTorch on the RX 7900 XTX Still Falls Short for Research
A measured look at where AMD ROCm with PyTorch and PyTorch Lightning still has rough edges on the RX 7900 XTX in 2026, and what that means if you are porting CUDA training workloads.
· 6 min read
GPT-5.5 Instant vs GPT-5.3: Which of OpenAI's Three Claims Hold Up
OpenAI swapped ChatGPT's default to GPT-5.5 Instant overnight, claiming faster responses, sharper reasoning, and fewer hallucinations. We grade each claim against independent testing and show developers what to change in their API stack.
· 6 min read
OpenAI Daybreak vs Anthropic Glasswing: Identical Benchmarks
Both shipped the same week with overlapping enterprise partners. What the convergence signals, and how to evaluate either for your AppSec pipeline.
· 6 min read
Macchiato Day 2 Review: Live Token Metrics and Parallel AI Terminals
Macchiato's Day 2 release ships a live token sidebar, per-agent cost dashboard, and shortcuts for Claude Code and OpenCode. Here is what changes for developers running multiple AI agents.
· 6 min read
Coolify Review: Self-Hosted Vercel Alternative on a $20 VPS
We deployed an Astro site, a Next.js app, and PostgreSQL on a Hetzner box, then compared this open-source PaaS to Vercel and Netlify on cost and reliability.
· 8 min read
OrbStack vs Colima vs Rancher Desktop: RAM and Startup
We ran all three as Docker Desktop replacements on macOS and Linux, and checked Docker Compose compatibility.
· 9 min read
PlanetScale vs Neon vs Supabase: Serverless DB in 2026
Compared on cold start latency, pricing at scale, branching workflows, and ORM compatibility -- and which one fits your architecture.
· 10 min read
SST Ion Review: TypeScript Compiled to Terraform
We migrated a production API from SST v2 to Ion and measured cold starts, deploy speed, and the live Lambda debugger.
· 9 min read
Turso Review: Edge-Native SQLite Built on libSQL
We measured query latency across 5 regions and compared Turso's primary-replica setup to Cloudflare D1 and Postgres. Plus where SQLite's limits matter.
· 8 min read
Cloudflare D1 Deep-Dive: SQLite at the Edge
How D1 delivers transactional SQLite with zero cold starts and read replication, its Workers integration, and where it fits among serverless databases.
· 8 min read
Fly.io vs Railway: Choosing a Modern PaaS for 2026
Compare regions, pricing, and developer experience across both platforms, and see which workloads each one handles best.
· 8 min read
Neon Review: Serverless Postgres With Storage/Compute Split
Database branching, serverless scaling, and a free tier - plus the cases where traditional Postgres still wins. A hands-on look.
· 7 min read
SST Ion Review: A Framework That Makes AWS Serverless Feel Coherent
SST Ion reimagines infrastructure-as-code by embedding AWS resource definitions directly into application code, with live Lambda debugging and a Terraform-compatible deployment engine. A review of the developer experience, the Pulumi migration, and where Ion fits in 2026.
· 8 min read
Supabase Edge Functions Review: Deno on the Edge
Runtime, developer experience, and integration with Postgres, auth, and storage, plus how it compares to Cloudflare Workers and AWS Lambda.
· 7 min read
Concurrency, Retries, and Timeouts in TypeScript AI Agents
Why Promise.race leaks model calls and billing, and how a single-owner pattern with AbortSignal, deadline budgets, and jittered retries fixes it.
· 6 min read
Temporal Hits 3,000 Customers: Durable Execution for Agents
Teams building long-running LLM agents are swapping DIY retry code for crash-proof workflows. What durable execution buys you, and what it costs.
· 6 min read
MinIO MemKV and the AI Recompute Tax
How MemKV offloads KV cache to persistent memory so agentic pipelines reload attention state, plus the 95% utilization claim and when reload beats recompute.
· 6 min read
Why AI Agents Fail Silently: Build an Observability Monitor
Agents return 200s and exit cleanly while hallucinating, degrading under rate limits, and overrunning budgets. Four failure modes, one minimal monitor.
· 6 min read
Why Long-Running AI Agents Break on HTTP
HTTP's request-response model drops connections mid-task. Ably's durable sessions keep messages, state, and reconnects intact.
· 6 min read
Querying Telemetry in Plain English
How the natural-language translation layer over logs, metrics, and traces works, what it genuinely helps with, and where it breaks.
· 7 min read
Caddy Web Server Review: Automatic HTTPS Without Ceremony
A close look at Caddy's Caddyfile syntax and reverse proxy setup, plus where it falls short compared to Nginx.
· 7 min read
Mac Mini as AI Agent Infrastructure
Apple Silicon's unified memory: benchmarks, real costs, Ollama and MLX setup, and honest tradeoffs versus cloud GPUs.
· 7 min read
NixOS & nixpkgs in 2026: Dev Environments Without Docker
How Nix flakes and devShells replace Docker for local dev: what works, where it hurts, and whether the learning curve is worth it for your team.
· 7 min read
The Rust Sidecar Pattern for Python AI Deployment
Python dominates ML development but struggles in production serving. Here's how the split works: Python handles models, Rust owns the hot path.
· 7 min read
SendGrid vs Mailgun vs Resend: 2026 Email API Comparison
An honest look at pricing, developer experience, deliverability, and fit to help you pick a transactional email API.
· 7 min read
SigNoz Review: An Open-Source Datadog Alternative
Unified logs, traces, and metrics on ClickHouse and OpenTelemetry. What it actually costs, and where self-hosting bites back.
· 7 min read
Streaming AI Inference: A Software Fix for LLM Energy Bills
Continuous batching, KV-cache management, speculative decoding, and model routing cut cost per token without new hardware.
· 7 min read
Temporal Cloud Serverless: Durable Execution
Run durable workflows on AWS Lambda with no infrastructure to manage. What changed, the tradeoffs, and when it fits.
· 7 min read
Stop Wrestling With Terraform State Imports at Scale
Config-driven import blocks and generated configuration replace the one-resource-at-a-time terraform import command, with a preview before you apply.
· 7 min read
Training an LLM in Swift: Matmul from Gflop/s to Tflop/s
How loop reordering, cache blocking, SIMD, multithreading, and GPU offload speed up matrix multiplication on Apple Silicon -- and why it sets training speed.
· 6 min read
Self-Hosting Guide on GitHub: Local LLMs and Home Servers
A review of mikeroyal's GitHub repo for WireGuard VPNs, Home Assistant, and private cloud -- plus where self-hosting saves money and where it doesn't.
· 6 min read
PyPI Package Growth Surge: What It Means for Python Devs
PyPI's catalog is growing faster than ever. Here's how that affects supply-chain risk and dependency bloat, and what to use when you audit your tree.
· 6 min read
Immich Review: Self-Hosted Google Photos Alternative
What it takes to deploy, how the mobile apps and on-device ML work, and the tradeoffs of hosting your own photos.
· 6 min read
Supabase Review: Dedicated Postgres for AI App Backends
The open-source Firebase alternative with auth, storage, realtime, and pgvector -- what holds up, and where pricing and the realtime engine bite.
· 6 min read
Coolify Review: Self-Hosted Vercel/Heroku Alternative
An open-source PaaS you self-host for about $6/month. We tested its 280+ one-click services to find where it beats Vercel and Heroku - and where it doesn't.
· 6 min read
rk3562deb: Can a $80 ARM Tablet Be a Linux Dev Workstation?
We read through the project that turns cheap RK3562 Android tablets into Debian machines: what works, what doesn't, and which dev workflows fit.
· 6 min read
70% of Americans Oppose Local AI Data Centers
The resulting permitting drag will hit inference pricing, region availability, and the architecture decisions developers make.
· 7 min read
Mozilla Defends VPNs to UK Regulators
Mozilla told Ofcom that VPNs are essential privacy infrastructure, not threats. Here's what changes for developers if regulators listen.
· 6 min read
Best Domain Registrars for Developers in 2026
Porkbun, Cloudflare, Namecheap, and Squarespace compared on API access, DNS management, WHOIS privacy, and renewal pricing. Stop overpaying on renewals.
· 7 min read
Deploying Bun Apps on Cloudflare Workers in 2026
Cold starts, free tier limits, the node:* compat story, and when Workers beats a VPS for side projects.
· 7 min read
Supabase vs Firebase in 2026 for Indie Developers
Postgres vs Firestore, auth, realtime, and pricing cliffs compared, plus when open-source ownership beats vendor convenience.
· 8 min read
Fly.io vs Railway: Fastest Side Project Deploy in 2026
We deployed the same Next.js app and Postgres to both, timing first deploy and cold starts. Railway won on speed; Fly.io won on global reach.
· 8 min read
Paperless-ngx: Self-Hosted Docs With a REST API
A hands-on review of the open-source DMS: Docker stack, OCR pipeline, AI workflow integration, and where Whoosh search hits its limits.
· 7 min read
Gitleaks: Open-Source Secret Scanning for Git Repos in 2026
Hands-on with the Gitleaks CLI, pre-commit hooks, and CI integration, plus how it compares to GitGuardian for teams that don't want per-developer pricing.
· 6 min read
Eleven Browser Games in a Week
All eleven live at play.pickuma.com. After the first two, the bottleneck was the chrome around each game, not game logic. The design system fixed that.
· 6 min read
I Shipped Two Web Games This Weekend
Stop at 7.77 and Eagle Run are live at play.pickuma.com: a 250-line vanilla canvas game and a one-button time-sense test, plus the stack and tradeoffs.
· 5 min read
Phantom Pulse RAT Hits Obsidian Community Plugins
The attack chain that put a RAT in developer vaults, plus how to audit plugins in Obsidian, VS Code, and Cursor.
· 6 min read
Best Free Tiers for Developers in 2026: SaaS, PaaS & IaaS
Which hosting, database, CI/CD, and observability free tiers still let you ship for $0, where the hidden cliffs are, and when paying beats the workarounds.
· 6 min read