Full-Stack AI Engineer

Building production-style AI + SaaS systems with FastAPI, Next.js, PostgreSQL, and LLM workflows — shipped as deployed MVPs.

I'm Rami Afif. I design and build AI and SaaS products end-to-end — from FastAPI / PostgreSQL backends to Next.js / TypeScript frontends — and ship them as deployed, clickable MVPs. Recent work includes a multi-tenant Storefront + POS SaaS and a full-stack AI productivity platform, alongside a set of focused AI demo apps.

2 Full-Stack Platforms AI + SaaS Portfolio FastAPI · Next.js · AWS

About Me

A developer who never stopped coding—from enterprise systems to AI platforms

Rami Afif
RA

Rami Afif

Full-Stack AI Engineer

Certified
Associate AI Engineer
DataCamp • Nov 2025
LinkedIn GitHub Email Me

The Real Story

I'm a Computer Engineering graduate (2009) pivoting back to tech after 12 years in construction project management. That background—coordinating complex schedules, resources, and workflows—shapes how I approach building systems today.

My software career started with 2 years as a Visual Basic Developer (2010-2012) at Saranay Recruitment, building enterprise hiring systems with VB + SQL Server. That's where I learned what production code means. During my construction years, I occasionally coded websites and trading bots as hobby projects—staying connected to programming but not at a professional level.

I started exploring AI in December 2022 when ChatGPT launched. In November 2024, I committed fully to becoming an AI Engineer. After 10 months of intensive study, I earned my Associate AI Engineer certification from DataCamp (November 2025).

Then I went into build mode. Over a focused build period, I built and shipped two full-stack platforms — a multi-tenant Storefront + POS SaaS and APE (an AI productivity engine) — plus a set of focused AI demo apps on Streamlit Cloud. These aren't tutorial copies; they're working systems solving real problems.

My platforms are production-style. My Storefront + POS SaaS uses FastAPI, a multi-tenant PostgreSQL design, and Redis with a Next.js / TypeScript frontend and AWS deployment services; APE pairs FastAPI + PostgreSQL + Redis + ChromaDB with multi-agent orchestration, async batch processing, and Docker-based deployment designed for low-latency responses under concurrent use. This is the kind of full-stack AI + SaaS work I want to do professionally.

Why Hire Me

Ships Independently

Takes features from idea to a deployed demo solo — I plan, build, and release.

Full-Stack Product Thinking

Builds complete flows — storefront, dashboard, POS, API, and database — not isolated scripts.

AI Used Responsibly

Uses AI tools seriously — but with human review, testing, and a proper Git workflow.

Idea → Deployed MVP

Turns an idea into a working, deployed MVP a hiring manager can click through.

Experience Timeline

Nov 2024 – Present

AI Engineer (Self-Employed)

Built and deployed two full-stack platforms — a multi-tenant Storefront + POS SaaS and APE (AI productivity engine) — plus a set of focused AI demo apps spanning RAG, multi-agent workflows, and GPU-accelerated ML.

2012 – 2024

Construction Project Coordinator + Hobbyist Developer

Dual track: Professional project management + Personal software development

Day job: Holyland Construction • Managed project scheduling, resource allocation, and budget tracking. Created daily/weekly/monthly work programs, material requirement planning (BOM), progress reports for contractors and clients.

After hours: Built websites for internet marketing, developed trading bots, experimented with new frameworks and tools—never stopped coding

2010 – 2012

Visual Basic Developer

Saranay Recruitment • Built enterprise hiring system with VB + SQL Server used by recruitment business

2009

B.S. Computer Engineering

Universidad De Zamboanga

🎯 Currently Seeking

Full-Stack AI Engineer / AI Engineer / AI Developer roles where I can build production-style AI + SaaS systems, work with LLMs, and ship features that matter. Remote-friendly, ready to start immediately.

Core Skills

AI / LLMs

RAG LangChain FAISS ChromaDB OpenAI Groq AWS Bedrock LLM workflows OCR / Textract

Backend

Python FastAPI Flask REST APIs Async processing JWT auth PostgreSQL Redis

Frontend

React Next.js TypeScript Streamlit Responsive UI

Cloud / DevOps

AWS ECS Fargate RDS S3 CloudFront WAF ALB / HTTPS Docker GitHub Actions Vercel

Featured Projects

Two full-stack platforms I designed, built, and deployed end-to-end.

More Projects

Focused AI demo apps showcasing diverse capabilities — RAG, LLMs, ML, and computer vision.

AI House Price Predictor

AI House Price Predictor

Oct–Nov 2025

Full-stack MLOps system with XGBoost trained on 4,162 homes achieving R² 0.52 and RMSE $198K. Flask API backend with Streamlit UI, dual-deployed for high availability.

Tech: XGBoost, Flask, Streamlit, Render.com

AI Movie Recommender

AI Movie Recommender

Nov 2025

RAG pipeline with Llama 3.2 + Chroma + Groq. Single-file Streamlit app with intelligent local/cloud fallback for 100% accuracy.

Tech: LangChain, Chroma, Groq, HuggingFace

AI Chatbot

AI Chatbot

Nov 2025

Real-time conversational assistant powered by Llama 3.1 via Groq. Features LangChain streaming with Ollama fallback for reliability.

Tech: LangChain, Groq, Llama 3.1, Streamlit

Zero-Shot Image Classifier

Zero-Shot Image Classifier

Nov 2025

Vision classifier using Llama 4 Scout (17B). Supports custom labels with structured JSON output in under 1.5 seconds.

Tech: Groq, Llama 4, LangChain, Streamlit

AI Resume Analyzer

AI Resume Analyzer

Nov 2025

RAG over PDF resumes with FAISS + Ollama. 100% local, private, and secure processing.

Tech: LangChain, FAISS, Ollama, PyPDF2

AI Job Matcher

AI Job Matcher

Nov 2025

GPU-accelerated resume-job matcher. Match score + AI edits + PDF report in under 2 seconds.

Tech: CUDA, HuggingFace, FAISS, fpdf

AI News Summarizer

AI News Summarizer

Nov 2025

Daily AI news digest with Llama 3.2 + embeddings. Zero-cost, offline-first architecture.

Tech: Ollama, FAISS, Streamlit, fpdf

Rami Afif AI Development Workflow v2.6 cover image

AI Development Workflow v2.6

June 2026

A human-governed Architect–Implementer Challenge system for building production-style AI and SaaS systems with structured challenge, independent diff review, and traceable verification.

Type: Engineering Workflow Article