Skip to content

About

I make large bodies of knowledge usable.

I'm Ali Abbas — a full-stack and AI engineer based in Melbourne, Australia. I like building the whole of a hard problem: the AI and retrieval at the core, the data infrastructure that feeds it, and the product people actually use. Today that's mostly knowledge systems; where I'm headed is broader — AI-powered products that turn messy, real-world information into software people can rely on, at real scale.

It started with a practical problem. The classical texts I wanted to read well weren't available in a form that was actually usable — searchable across languages, readable on a phone, and faithful to the source. So I built one. Shia Library began as a reading app and became a systems project the moment I tried to make Arabic search actuallywork, keep tens of thousands of pages fast, and translate a library's worth of text without dropping or fabricating a single word.

Each hard part pulled me one layer deeper. Search relevance led to diacritic-insensitive indexing and hybrid retrieval. Serving a large read-mostly corpus led to a three-layer caching strategy and row-level access control. Translating the corpus led to a resilient, resumable LLM pipeline with real quality gates. The pieces add up to a single platform I own end-to-end — and to a clear sense of the problems I want to keep working on.

How I work

Depth over breadth.I'd rather operate one real system and understand its hard parts than ship ten demos. Measure, don't assert. If I claim a number, I want it reproducible. Honest about limits. Every project here lists what I'd do differently — knowing where something breaks is part of building it. Build for the reader. The engineering exists to make something genuinely useful, not the other way around.

Currently

Finishing my degree at Monash while deepening the platform — citation-grounded retrieval and an evaluation harness — and deliberately branching into new problem spaces (developer tooling, distributed systems, applied AI) to broaden my range. Open to software, AI, and platform engineering roles where these problems are the point.

Capabilities

Languages

  • TypeScript
  • JavaScript
  • Python
  • Java
  • SQL

Frontend

  • React
  • Next.js (App Router)
  • Tailwind / MUI
  • Accessibility
  • Performance

Backend & data

  • Node.js
  • PostgreSQL
  • Supabase
  • pgvector
  • Full-text search
  • Row-level security
  • Caching

AI & retrieval

  • RAG
  • Embeddings
  • Hybrid & semantic search
  • Prompt engineering
  • LLM evaluation
  • Prompt caching
  • Anthropic / OpenAI APIs

Infra & practices

  • Vercel
  • CI/CD (GitHub Actions)
  • Testing (Vitest, Playwright)
  • Git
  • Security (CSP, secret scanning)