Prompt caching is a prefix problem
Prompt caching reads cost a tenth of input on most tiers. The prefix that qualifies runs 512 to 4,096 tokens, and that floor picks your cheapest model.
Articles
Blog
Practical notes on building, deploying, and shipping real products for solo devs and small teams.
AI layoffs are inversion of control
AI layoffs are sold as obsolescence. The quotes, the labor data, and the hiring records point at bargaining power. How to tell the difference.
AI code review is not independent
AI code review is not an independent check: two AI passes miss the same bugs. A missed CI injection, an npm worm in repo config, and swarm data show it.
AI model fatigue is a business rhythm
AI model fatigue is a business rhythm. The deltas are small, the cadence serves the labs, not you. Own your workflow, gate every switch on your own evals.
MiniMax H3 for local video
MiniMax H3 ships open weights for video generation with native stereo audio. What you actually get running it on your own GPU, and what stays in the cloud.
The full disk that killed my database
A full disk corrupted my database's write-ahead log and it wouldn't start. The recovery runbook, and ops habits I deferred until Saturday made them due.
Read the code AI writes
An AI agent wrote 400 lines that passed every test. Nobody could read them. A 30-minute human rewrite beat it. Here is why hard skills still matter.
Qwen3.8 Max for builders
Qwen3.8-Max is Alibaba's 2.4T MoE flagship - #2 in Vision Arena, $2/$6 per million tokens, open weights next week. The catches that decide.
Kimi K3 for builders who ship
Kimi K3 is the largest open model, fourth on the independent index. The verified numbers, and the catches that decide whether you switch your coding agent.
Ad attribution without a data team
You do not need a marketing data platform to know which ad drove a sale. Here is attribution with UTM tags, one landing page, and your database.
TLS without losing sleep
TLS breaks on a VPS two ways: silent expiry and deploys that drop sessions. Set it up so renewals run, sessions survive, and you stop worrying about it.
App Store screenshots that convert
Screenshots are the product page element that drives install. Here is a screenshot test workflow for indies: lead with one benefit, test, read the result.
Measure what matters when you ship solo
Most dashboards for solo products are vanity arrays. Here are the few numbers that decide whether you survive, and the metrics that look important but lie.
Pick a stack you can run alone
For a solo builder, the best stack is the one you can debug in a crisis and afford at low revenue. Boring tech wins when you are the only operator.
One-time or subscription pricing
Subscription compounds revenue but adds churn, trust debt, and operational weight. One-time is simpler but caps growth. Here is how to choose.
Your first 100 dollars on ads
A sober first-budget ad playbook for solo builders: one audience, one platform, one offer, kill losers fast. Built to learn, not to scale.
ASO keyword workflow with Astro
A hands-on ASO workflow for solo iOS developers using the Astro tool: find keywords that fit 100 characters, track rank, and update without guessing.
From AI prototype to paid product
AI gets you a prototype in minutes. The real work is webhooks, deploys, TLS, and refunds. Here is what lives in the gap and why it decides whether you ship.