Posts
Page 8 of 66
-
API vs Self-Hosted LLMs: The Real Cost
API vs self-hosted LLM cost reality, GPU TCO, privacy, latency, break-even math. When paying OpenAI/Anthropic wins. When local wins.
10 min read -
Free AI Image Gen vs Your Own GPU
Free AI image generators win on convenience but lose fast on volume, style consistency, and control. Here's the real break-even against running your own GPU.
12 min read -
RAG Beyond Vector Search: BM25, Hybrid, Re-ranking
RAG beyond dense embeddings: BM25 lexical search, hybrid retrieval (RRF, weighted fusion), cross-encoder re-rankers (BGE, Cohere, Jina). When pure vector search fails.
11 min read -
Kaniko & Buildah: Daemonless Container Builds
Daemonless OCI image builds in CI without Docker-in-Docker risks. Kaniko vs Buildah compared, when each fits, and which one suits your pipeline.
10 min read -
Podman Compose for Docker-Free Workflows
Drop Docker's daemon entirely. Podman rootless containers with podman-compose, drop-in compatibility, daemonless orchestration, and where it bites you.
9 min read -
LXC vs Docker: Lightweight Containers
LXC and Docker both containerize workloads, but one runs mini-VMs with init processes while the other handles single-process apps. Here's when each wins.
10 min read -
How to Stretch a Free LLM Tier
Free LLM tiers usually die from bloated prompts, not stingy quotas. Context discipline, caching, tiering, batching, and backoff that triple your runway.
12 min read -
3-2-1-1-0: The Backup Strategy That Survives Ransomware
3-2-1 backups aren't enough anymore. The 3-2-1-1-0 rule adds immutable & offline copies plus verified restores, here's how to implement it.
9 min read -
regclient: Container Image Lifecycle Without a Daemon
regctl, regbot, and regsync give you tag retention policies, registry replication, and manifest surgery, no Docker daemon required.
9 min read -
SSH Bastion & Jump Host Patterns That Don't Hurt
Stop opening port 22 to the world. SSH bastion hosts, ProxyJump chains, session recording, and self-hosted Teleport alternatives that actually work.
9 min read -
Linux Real-Time Kernel: When PREEMPT_RT Actually Earns Its Keep
PREEMPT_RT hit Linux mainline in 2024. Here's what real-time actually means, when you need it, and when it'll just slow you down.
9 min read -
The Free AI Stack: What $0 Gets You
A capability map of what's free from AI providers in 2026: chat, APIs, image generation, speech tools, and coding help, with the real quotas.
13 min read
