Tag: deepdive

13 articles

· ai-tools / llm

Context engineering in 2026 — six patterns that work

Context engineering decides what your model sees at inference. Six patterns with code: ordering, caching, compaction, sub-agent isolation, and more.

· ai-tools / code-review

AI code review tools 2026: 7 tools tested on real bugs

CodeRabbit leads by F1 score (51.2%, Martian). Qodo Merge is top for self-hosted. Snyk Code wins on security. What each tool actually catches — and misses.

· cursor / rules

Cursor rules: Four modes and the .cursorrules trap

.cursorrules silently fails in Agent mode — 0/9 compliance. The four rule modes that work, five silent anti-patterns, and six annotated real-world templates.

· ai-tools / llm

Prompt caching in 2026 — Anthropic, OpenAI, and Gemini compared

Prompt caching cuts costs 90%. Anthropic requires explicit markers, OpenAI caches automatically, Gemini bills hourly. Here is which one fits your workload.

· testing / vitest

The test pyramid is dead — what replaced it

The 70/20/10 ratio nobody official ever endorsed, how Vitest browser mode and Playwright component testing dissolved the tier boundaries, and a concrete testing strategy for 2026.

· ci / docker

Build cache optimization: what actually works in 2025

Your CI is slow because of wrong cache layer, mode, or key — not hardware. Systematic fixes for Docker BuildKit, GitHub Actions, and Turborepo remote cache.

· feature-flags / architecture

Feature flags as architecture — the long-term cost

Feature flags cut shipping risk on day one. Left unmanaged, they become invisible operational debt. Here is how drift compounds and how to stop it.

· stripe / paddle

Stripe Tax vs Paddle MoR: real-world fee math for SaaS

Paddle wins below $10k MRR regardless of geography. Stripe pulls ahead above $30k MRR if your customers are US-heavy. Exact fee math and a decision matrix.

· supabase / postgresql

Supabase RLS pitfalls — what we learned in production

7 Supabase RLS pitfalls: NULL auth.uid(), service_role leaks, 171ms policies, infinite recursion, privilege escalation — with the exact SQL fix for each.

· monorepo / turborepo

Turborepo monorepo pitfalls we learned the hard way

5 production pitfalls that consistently bite mid-size teams 3–12 months into a Turborepo + pnpm monorepo migration — and exactly how to fix each one.

· typescript / strictness

Does TypeScript strictness actually reduce your bug rate?

TypeScript strict mode catches ~15% of committed bugs — not the 38% from Airbnb. The honest verdict from peer-reviewed data and 3 industry case studies.

· cloudflare / aws

Cloudflare vs AWS: the complete cost breakdown at scale

Cloudflare wins most mid-scale workloads; AWS wins large databases. Real cost breakdown: Workers vs Lambda, R2 vs S3, D1 vs RDS with exact figures.

· ai-agents / costs

The real cost of running an AI agent team in 2026

API bills are the smallest number in your AI agent budget. Solo founder: $195/month, $1,470 TCO. 10-engineer startup: $2,440 out-of-pocket, $8,740 true TCO.