The blog.

Deep dives, comparisons, and roundups on the AI tools worth paying for. Only what we'd actually buy — no SEO sludge, no influencer demos.

173 posts
Review: Bolt.new review: the best build, the worst reviews

Bolt.new review: the best build, the worst reviews

Bolt.new review from a measured build: the same six-requirement prompt I gave Lovable and Replit, 100,000 tokens, and the 1.4-star gap nobody explains.

Review: Lovable review: six requirements, six passes, 109 seconds

Lovable review: six requirements, six passes, 109 seconds

Lovable review from a real build. One prompt, six requirements, a working app in 109 seconds — and an honest look at what it got right and where it stops.

Guide: AI tool cost statistics: 60+ figures, each labelled by source

AI tool cost statistics: 60+ figures, each labelled by source

What AI video, voice, coding and automation tools really cost — per clip, per call, per invoice. 60+ figures from our own accounts, each labelled by source.

Roundup: Cursor alternatives: I left after $510.70, here's where I went

Cursor alternatives: I left after $510.70, here's where I went

Cursor alternatives from someone who actually cancelled. Real invoices, what each option does about the metered bill, and which ones I've tested.

Roundup: Claude Code alternatives: 7 tools, and what switching costs

Claude Code alternatives: 7 tools, and what switching costs

Claude Code alternatives ranked by what they actually cost. Real invoices, the vendors' own $150-250 figures, and which ones I've tested versus only read about.

Guide: Codex pricing: what you actually pay in 2026

Codex pricing: what you actually pay in 2026

Codex pricing runs $0–$200/mo inside a ChatGPT plan, plus token-priced credits past your limit. A credit is four cents. What it really costs in 2026.

Comparison: Codex vs Claude Code: I quit Codex. Then I retested it.

Codex vs Claude Code: I quit Codex. Then I retested it.

Codex vs Claude Code from someone who dropped Codex in May over one specific flaw — and found it fixed in July. What changed, and what I'd still watch.

Comparison: Cursor vs Codex: I paid for one and quit the other

Cursor vs Codex: I paid for one and quit the other

Cursor vs Codex from someone who cancelled Cursor after $510.70 and walked away from Codex too. What each actually costs, and which one I'd go back to.

Comparison: Claude Code vs Cursor: I paid for both for a year

Claude Code vs Cursor: I paid for both for a year

Claude Code vs Cursor, decided on invoices rather than vibes. They have converged on capability. Three axes still separate them, and only one decided it for me.

Review: Cursor review: I paid for 11 months, then cancelled

Cursor review: I paid for 11 months, then cancelled

Cursor review from someone who actually paid: $510.70 across 11 months, a third of it metered usage. What it does well, and what I got wrong about the bill.

Guide: Claude Code pricing: what it costs and what it really costs

Claude Code pricing: what it costs and what it really costs

Claude Code pricing is $20, $100 or $200 a month. Anthropic's own docs put enterprise usage at $150–250 per developer per month. Here's why the gap matters.

Guide: How to use Claude Code: setup, then the part that matters

How to use Claude Code: setup, then the part that matters

How to use Claude Code, from the install command to the habits that decide whether it works. Every best practice descends from one constraint most guides skip.