Skip to content
View AJSubrizi's full-sized avatar

Block or report AJSubrizi

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
AJSubrizi/README.md

Antonio Junior Subrizi

Full-Stack Engineer & AI-Augmented Builder
Founder & Lead Engineer @ NailDash • Self-taught

“Simplicity is the ultimate sophistication.”


I build real products end-to-end: frontend, backend, distributed systems, and AI applications.
Self-taught, with a focus on shipping production-grade software that solves actual business problems.

Mindset: ship fast, think clearly, build things that work.

What I Do

  • Design and ship full-stack web apps and SaaS platforms (Next.js, React, TypeScript, Tailwind)
  • Build backends, APIs, databases and auth flows with strong emphasis on security and scalability
  • Deliver to production with Docker, Vercel, AWS and CI/CD
  • Use AI as a real engineering multiplier — agents, automation, tooling and workflows
  • Focus on product quality, security, UX, maintainability and execution

🧰 Tech Stack

Languages
TypeScript JavaScript Python Go Rust PHP C++ Java C# HTML5

Frontend
React Next.js Svelte Vue.js Tailwind CSS shadcn/ui Vite Three.js React Native

Backend & APIs
Node.js FastAPI Express Django GraphQL Prisma

Databases & Data
PostgreSQL Supabase MySQL MongoDB Redis Firebase

Cloud, DevOps & Infrastructure
Docker Vercel AWS Kubernetes Google Cloud Linux Git CI/CD Jest Playwright

AI, Agents & Automation
CrewAI OpenAI API TensorFlow RAG systems AI Agents Prompt Engineering n8n

Tools
Figma Bash GitHub n8n

🚀 Featured Projects

Project Description Stack
NailDash Multi-tenant SaaS for nail studios and beauty businesses React, TypeScript, Supabase (RLS), Paddle, Playwright
Mancin Coffee & Wine Full-stack commerce + operations platform with heavy automation Next.js, Prisma, PostgreSQL, AI recs, Telegram + n8n

🤖 AI & Agent Tooling (Open Source)

Project Description
Synapse Persistent, Obsidian-compatible memory for AI coding agents (Claude, Cursor, Gemini...). Shell-native, no database.
promptcrash Local-first crash reporter for LLM applications with PII redaction and regression testing
scratch-to-prompt Turn messy ideas into high-quality, agent-ready prompts
AgentsMD.space Public directory of AGENTS.md and CLAUDE.md files for coding agents

🔬 Research

  • HRM-ReRead-MT — PyTorch research scaffold for hierarchical re-reading, latent memory consolidation and teacher-guided distillation

Experience

  • Founder & Lead Engineer — NailDash.it (2026–Present)
    Building a multi-tenant SaaS platform with row-level security, recurring billing and production-grade architecture.

  • Full-Stack Developer & AI Architect — Mancin Coffee & Wine (2025)
    End-to-end platform with automation, AI recommendations and integrated operations.

  • Full-Stack & Automation Developer — Freelance (2023–Present)
    Custom software, n8n automations, CrewAI agents and RAG systems for real businesses.

Philosophy

Modern developers shouldn’t only write code. They should understand the product, the user, the business logic, the architecture, the risks, and the execution.
Code is only valuable when it solves a real problem.


→ ajsubrizi.dev · @AJSubrizi

Pinned Loading

  1. synapse synapse Public

    Persistent memory for AI coding agents give Claude Code, Codex, Gemini & OpenCode a shared, Obsidian-compatible knowledge base. Open-source, shell-native, no database.

    Python 3 1

  2. OhCanvas OhCanvas Public

    Infinite canvas with real PTY terminals, browser nodes, drawing tools, and AI agent control — built with Tauri 2 + React + Konva

    TypeScript

  3. scratch-to-prompt scratch-to-prompt Public

    Turn messy ideas into agent-ready prompts. A small skill for Claude Code, Cursor, built on Anthropic’s prompt-engineering principles.

    Shell 2 1

  4. HRM-ReRead-MT HRM-ReRead-MT Public

    Research scaffold for hierarchical re-reading, latent memory consolidation, and teacher-guided distillation in PyTorch.

    Python

  5. agentsmd.space agentsmd.space Public

    A small directory of AGENTS.md and CLAUDE.md files for AI coding agents.

    TypeScript 1