Recent articles
Insights on modern web development, performance, and building better digital products.
Claude Fable 5 Makes Codebase Audits a Cost-Control Problem
Claude Fable 5 was briefly available before Anthropic's June 12 suspension notice. The article remains useful as a planning pattern for AI codebase audits.7 min readHermes Desktop: Personal Agents, Operational
Hermes Desktop gives Hermes a real personal agent runtime: profiles, remote gateways, Portal setup, visible state, and policy boundaries for always-on work.9 min readAI Readiness Assessment: The 10-Point Checklist webvise Scores Before Any Build
An AI readiness assessment scores owners, sources of truth, and written decisions, and none of its ten points ask about models. Here is the checklist webvise scores prospects against, with a real assessment arc.6 min readHow Long Does It Take to Build an MVP? A Practical 3 to 5 Week Plan
A focused MVP takes 3 to 5 weeks when it tests one workflow with standard auth, one data model, and one success metric. Here is the honest timeline and what stretches it.8 min readMVP Requirements Document Template: The Scope That Ships
A good MVP requirements document is not a feature backlog. Use this template to define the learning goal, cut scope, and brief a build that can ship.8 min readAI Consulting for Small Business in 2026: What a Real Assessment Delivers
AI consulting for small business should end in a working prototype, a written DSGVO answer, and a build or no-build decision. This is the deliverable list webvise publishes and holds itself to.5 min readHow to Calculate AI Automation ROI: Time One Document, Price One Run
AI automation ROI is measured in one afternoon: time a real document through the old path and the new path, price the run, and label every input you did not measure. With the stopwatch numbers from a real construction client.7 min readHermes Profiles: When to Split Your Assistant
Hermes profiles separate memory, sessions, skills, identity, cron state, and gateway routes so recurring AI roles can stay narrow, auditable, and useful.9 min readYour AGENTS.md Is Your Moat Now: What Goes Inside Production Files
Next.js 16.2 made AGENTS.md default scaffolding on Tuesday. Augment's empirical study and my production template show what useful agent instructions actually contain.9 min read