Production discipline
Auth, migrations, backups, CI, and deploy guardrails are part of the work, not afterthoughts.
Software developer · AI-first systems
Building a personal AI-first hub as engineering evidence.
I am turning a real single-user productivity system into a public case study: modular Go backend, TypeScript frontend, production deploys, and an agent workflow where specifications, boundaries, and human approval gates are first-class.
Public site CV, projects, blog, and privacy-safe case studies on anton415.ru.
Go API Modular monolith with inward dependencies.
React hub Private app surface behind Yandex ID.
PostgreSQL Persistent module data with migrations.
AI workflow ChatGPT for scope, Claude Code for implementation, Codex for visual polish.
What I am building
A production-oriented personal system that is useful day-to-day and reviewable as a portfolio artifact.
AI-readable architecture
Small files, explicit layers, and enforced dependency direction make agent edits safer.
Honest public boundary
The portfolio explains the system without exposing private tasks, finances, or notes.
Featured project
The central case study for the repository and workflow.
AI-first Hub
In ProgressA personal modular hub for productivity, finance, automation, and AI-assisted engineering — built as a production-oriented single-user system and as a training ground for AI-first software development.
Latest writing
Engineering notes tied to real shipped changes and decisions.
-
Building a Personal AI-first Hub as a Path to AI Software Engineer
Why I am building a single-user modular monolith as both a useful personal product and a deliberate training ground for AI-assisted engineering — and why the goal is practical engineering ability, not chasing AI hype.