Amyy AI
Architected Amyy AI — a privacy-first AI companion serving 300+ users with v2 backend, OAuth, rate limiting, and LLM workflows shipped end-to-end.
Amyy AI is a privacy-first AI companion framework serving 300+ real-world users. I architected the v2 backend infrastructure — designing the service layer, data models, and API contracts — and built the major frontend, integrating FastAPI services with the client and shipping with Claude Code for fast delivery.
Project At A Glance
Timeline
May 2026
Industry
AI · SaaS · Privacy
Contribution
Backend infrastructure, API contracts, OAuth, rate limiting, caching, FastAPI integration, and frontend delivery
Collaboration
Project Visual
Amyy AI
Summary
What this project demanded.
Amyy AI is a privacy-first AI companion framework serving 300+ real-world users. I architected the v2 backend infrastructure — designing the service layer, data models, and API contracts — and built the major frontend, integrating FastAPI services with the client and shipping with Claude Code for fast delivery.
Role
Full Stack Developer
Year
2026
Problem Space
Sustaining concurrent traffic for an AI companion app meant solving auth, rate limiting, caching, and database tuning together. The product also had to feel calm and trustworthy on the surface while sitting on top of LLM workflows and a multi-service backend.
I implemented Google OAuth 2.0, sliding-window rate limiting, response caching, and tuned database queries to hold up under load, plus CI/CD pipelines for deploying on AWS Amplify with Docker. The frontend had to keep technical complexity invisible to users while staying integrated with FastAPI endpoints.
Capability 01
FastAPI
Capability 02
Next.js
Capability 03
PostgreSQL
Capability 04
Google OAuth 2.0
Capability 05
AWS Amplify
Capability 06
Docker
Capability 07
REST APIs
Capability 08
CI/CD
Context
Amyy AI needed a real backend before it could become a real product.
The brief was to take an AI companion concept and ship a production-grade v2 with the systems work that real users require — auth, rate limiting, caching, query tuning, and CI/CD — while still delivering a frontend that felt calm and trustworthy.
01
What I owned
- Architected v2 backend infrastructure: service layer, data models, and API contracts
- Implemented Google OAuth 2.0, sliding-window rate limiting, and response caching
- Tuned database queries and set up CI/CD pipelines for AWS Amplify + Docker
02
What it powered
- Concurrent traffic for 300+ real-world users on a privacy-first companion app
- FastAPI service integration with the client and the major user-facing frontend
- Fast delivery cycles with Claude Code, shipping backend and frontend together
Signal
300+
Users served by the v2 backend infrastructure on a real-world AI companion app
Signal
OAuth 2.0
Google OAuth 2.0, sliding-window rate limiting, and response caching shipped to production
Signal
AWS + Docker
CI/CD pipelines deploying the platform on AWS Amplify with Docker
Production reliability was the product feature.
Insight
Privacy-first AI products live or die on the boring systems: auth, rate limits, caching, and pipelines. Get those right and the surface above can stay calm.
Core Interaction Shifts
The product decisions that changed how the experience felt.
Shift 01
v2 Backend Architecture
Designed the service layer, data models, and API contracts powering chat, auth, and core companion flows for 300+ users.
Shift 02
Production-Grade Reliability
Implemented Google OAuth 2.0, sliding-window rate limiting, response caching, and query tuning so the platform sustained concurrent traffic under load.
Shift 03
Frontend + FastAPI Integration
Integrated FastAPI services with the client and built the major frontend; shipped with Claude Code for fast delivery and tight iteration cycles.
Influence & Validation
What changed because of the work.
Amyy AI moved from prototype to a production-ready product with the auth, rate limiting, and CI/CD foundations needed to scale to 300+ users.
v2 backend now powers real-world traffic with OAuth, rate limiting, and Redis-backed caching in place
AWS Amplify + Docker CI/CD pipeline shipped, enabling fast and safe production deploys
Frontend-FastAPI integration delivered the major user-facing surface alongside backend infrastructure