Annus Shabbir
Available for freelance & contract work

AI Engineer · LLM/RAG · Full-Stack Developer

I build production AI & RAG systems — and the web apps around them.

AI engineer and full-stack developer with 5+ years shipping to production and an MS in Artificial Intelligence from LUMS. I design and ship AI/RAG platforms, AI chatbots, and fast, SEO-friendly web apps.

Lahore, Pakistan
Who I help

Built for teams shipping real things.

Founders, businesses, and teams who need AI features, products, or sites that actually ship.

Startups & founders

Ship an MVP or add an AI feature — fast, without cutting corners.

Businesses with knowledge

A grounded AI chatbot or assistant over your docs, data, and support content.

Teams needing AI & automation

RAG systems, agents, automation workflows, and backend/API work.

Anyone needing a great site

A fast, SEO-friendly business website that actually converts.

01 — Services

What I can build for you.

From a focused AI feature to a full product — here's where I help most.

AI / LLM Solutions

Production RAG systems, AI chatbots, and agentic workflows grounded in your data — with citations, streaming responses, and evaluation built in.

  • RAG
  • Chatbots
  • Agentic AI
  • MCP
  • LangGraph
  • Vercel AI SDK

ML & NLP Pipelines

Custom ML/NLP pipelines — model optimization and edge deployment, speech and NLP workflows, and rigorous LLM evaluation.

  • PyTorch
  • Hugging Face
  • Whisper
  • Quantization
  • LLM Evaluation

Full-Stack Web Development

Fast, SEO-friendly web apps and business sites — from MVP to production, built to be maintainable and to last.

  • Next.js
  • React
  • Node.js
  • Django
  • Tailwind CSS

Backend & API Engineering

Scalable backends and APIs — FastAPI and Node services, real-time and event-driven systems, and cloud infrastructure on AWS.

  • FastAPI
  • Node.js
  • PostgreSQL
  • Redis
  • AWS
02 — Selected work

Production systems, shipped.

Government-scale AI services, RAG platforms, and full-stack products — with the metrics behind them.

Case study · 2025

AskTT — Government AI Service

Phaedra Solutions

Production FastAPI RAG backend for the MyTT government portal — hybrid retrieval, cited answers, streaming + voice chat, plus an LLM recruitment-matching engine.

400,000+applicant–vacancy pairs evaluated
30–50/secreal-time matches
  • FastAPI
  • Elasticsearch
  • RAG
  • OpenAI
Case study · 2021–2025

Mentalhealth Match

Esketchers

A therapist-matching platform serving 30,000+ therapists — rebuilt on Next.js 14 with Angular SSR, reaching 85+ PageSpeed and stronger SEO.

85+Google PageSpeed on key pages
30,000+therapists supported
  • Next.js 14
  • Angular SSR
  • Node.js
  • PostgreSQL
Research · 2025

LLM Optimization for Edge Deployment

Compressed LLaMA-3.1-8B by 73.6% (15GB → 3.95GB) with a 273% CPU decode speedup — running on a Raspberry Pi 4.

73.6%model size reduction (15GB → 3.95GB)
273%CPU decode speedup (1.16 → 4.33 tok/s)
  • PyTorch
  • llama.cpp
  • GGUF
  • CUDA
How I work

A few things you can count on.

01

Clear communication

You always know what's happening, why, and what's next.

02

Production-grade

Maintainable, tested, and built to last — not a throwaway prototype.

03

Ships fast

Tight feedback loops and working software early and often.

04

Software + modern AI

I bridge solid engineering with modern AI, so the AI actually ships.

Have a project in mind?

Tell me what you're building. I'll reply with how I'd approach it.

Start a conversation