SELECTED WORK

AI engineering & data science projects

Products and experiments by Arzuman Abbasov, Data Scientist & AI Engineer.

Professional · Spice Media Production

Spice Studio ↗

A professional client project I built for Spice Media Production. Their fashion photography workflow involved physical studio sets, lighting rigs, models and garment preparation. Spice Studio translates that production discipline into a guided AI workflow for fashion catalog imagery.

Next.js, TypeScript, Python, Flask, Gemini, Topaz, SQLite

Voice-powered AI tutor · ElevenHacks #10 · 2nd place

Claw Learn ↗

Make abstract mathematics easier to understand through conversation and visual explanations. Claw Learn lets a learner ask a math or physics question and explore the answer through live animation.

Next.js, React, TypeScript, ElevenLabs Speech Engine, WebRTC, Canvas, Tailwind CSS

AI & global perspectives · Perplexity Hackathon · Most Fun / Creative Project

What Does the World Think? ↗

Explore how the same claim may be interpreted across cultures. The project turns a statement into AI-generated perspectives for 192 country contexts, revealing where agreement and disagreement may emerge.

React, TypeScript, Perplexity Sonar, Leaflet, Tailwind CSS, shadcn/ui

Open source · Browser extension

Quiet X ↗

An open-source Chrome extension for a quieter X timeline. It hides marked advertisements and posts matching user-selected account-country, language and topic rules without inserting replacement cards.

TypeScript, Chrome Manifest V3, Node.js, TypeSafe AI

Hackathon · Generative media · Backblaze Generative Media Hackathon

BrandBlaze ↗

A product campaign tool built for the Backblaze Generative Media Hackathon. One reference photograph becomes campaign variants for different markets, channels and environments, with the product’s identity treated as a constraint.

Python, FastAPI, React, Claude, Genblaze, GMI Cloud, Backblaze B2

Hackathon winner · Accessible education · Teknofest 2024 · T3 AI Hackathon · 1st place

Spark AI ↗

A team-built learning platform designed around different accessibility needs. Spark explores personalized learning materials for dyslexia, ADHD and color blindness, alongside virtual laboratory experiences for learners without physical lab access.

Python, Django, React, PostgreSQL, Redis, Celery, FFmpeg