Skip to content
z.dev

Case study — 2025 — 2026

Ayei.ai

Custom-built AI employees, ready in days. A service platform offering custom-built AI employees — sales, marketing, bookkeeping, research, reception, and more — each tailored to the client's exact workflow and running 24/7.

Type
saas
Years
2025 — 2026
Stack
Next.js · Node.js · Python · OpenAI API · PostgreSQL · Automation APIs
Ayei.ai — live product interface
Captured from the live productVisit live

The problem

Small teams drown in repetitive knowledge work but can't justify full-time hires for every function. Generic AI tools don't fit their workflows; custom automation is usually a consulting-priced project.

My role

Core engineer: the AI-assistant runtime, per-client workflow customization, and the delivery pipeline that gets an assistant from intake to production in days.

The hard part

Every client's workflow is different, so the platform has to assemble assistants from reusable capability blocks — monitoring, outreach, reporting, bookkeeping — while keeping each deployment isolated, auditable, and cheap enough to run around the clock. The interesting engineering is in the assembly line, not any single bot.

Outcome

Live at ayei.ai with eleven assistant roles offered — from lead qualification to executive assistant — each custom-built per client and deployed within days.

Ayei.ai — product features view
Ayei.ai — mobile interface
Deeper views of the live product — features page and the 390px mobile viewport. Responsive behavior is part of the build, not an afterthought.