Loading
About Work Experience Skills Achievements Testimonials Services Contact Book Meeting Download Resume

Hello, I'm

Noor Aftab

|

Engineering Intelligence into Infrastructure — Where AI Meets Automation at Scale

View My Work Get In Touch
0+ Years Experience
0+ Projects Completed
0+ Happy Clients
Scroll
01

About Me

MLOps / ML Platform Engineer with 4+ years of experience building and operating production-grade AI platforms at scale. Proven track record in end-to-end MLOps pipeline design, GPU infrastructure management, and inference optimization. Orchestrated model serving across 60+ Kubernetes clusters processing 100,000+ daily requests. Achieved 60% infrastructure cost reduction through GPU consolidation and shared inference architectures.

📍 Lahore, Pakistan
✉️ Nooraftab779@gmail.com
Download Resume
Noor Aftab
02

Featured Work

A selection of projects I've worked on

ShuftiPro - Enterprise AI & MLOps Platform
MLOps & Enterprise AI

ShuftiPro - Enterprise AI & MLOps Platform

Large-scale KYC and document intelligence platform supporting 100K+ daily requests across 60+ Kubernetes clusters with AI-driven identity verification.

YOLO MRCNN PaddleOCR ONNX vLLM Gemini GPT LLaMA Qwen Python FastAPI Cython Kubernetes Docker Elasticsearch MongoDB PostgreSQL
Shein AI Modesty Classification System
AI/ML & Web Scraping

Shein AI Modesty Classification System

Production-grade system that scrapes Shein's clothing catalog, extracts product metadata, and classifies images using AI-powered modesty detection.

Python Selenium PostgreSQL NudeNet Detectron2 DensePose FastAPI Pandas
ZebraInvest Real Estate Intelligence
AI/ML & Data Engineering

ZebraInvest Real Estate Intelligence

Large-scale UK property intelligence system with automated crawling, price tracking, and AI-enhanced OCR for energy performance certificate extraction.

Python Selenium PostgreSQL YOLO pytesseract OpenCV Rotating Proxies
Careerli - AI Resume & Cover Letter Platform
AI/ML & Serverless

Careerli - AI Resume & Cover Letter Platform

Enterprise-grade AI platform for automated resume writing, cover letter generation using GPT-4 with sophisticated prompt engineering.

Python OpenAI GPT-4 AWS Lambda FastAPI Pydantic REST API
Qubyk AI Product Categorization
AI/ML & Computer Vision

Qubyk AI Product Categorization

Enterprise AI system using Vision-Language Models to automatically classify products into hierarchical taxonomy for Australian e-commerce marketplace.

Python PyTorch IDEFICS-9B LoRA RabbitMQ Hugging Face CUDA
ASX Announcement Intelligence Pipeline
AI/ML & Data Engineering

ASX Announcement Intelligence Pipeline

Serverless AI-powered system for extracting and structuring financial announcements from Australian Stock Exchange using Claude 3 Haiku.

Python Azure Functions Claude AI PyPDF2 pdfplumber Pandas Docker
Floorplan Analysis System
Computer Vision & OCR

Floorplan Analysis System

Automated system for extracting structured geometric and semantic information from architectural floorplans using YOLO and OCR.

Python FastAPI PyTorch YOLOv8 PaddleOCR OpenCV MongoDB PostgreSQL Docker
03

Experience

My professional journey

Lead AI / ML Engineer – MLOps & Platform Architecture

ShuftiPro
Nov 2022 — Present Remote Current Full-time

Lead the design and operation of a distributed AI platform for KYC, identity verification, and document intelligence, processing 100,000+ verification requests per day across 60+ Kubernetes clusters.

  • Architected end-to-end MLOps pipelines — data ingestion, model inference (OCR, computer vision, LLM extraction), evaluation, and monitoring — with confidence-based fallback and validation mechanisms
  • Re-architected LLM inference using vLLM-based shared GPU services with continuous batching and optimized VRAM allocation, reducing GPU consumption from 60+ GPUs to 3–5 shared GPUs
  • Achieved 60% infrastructure cost reduction, lowering monthly GPU/compute expenses from USD 20,000 to USD 7,500 through GPU consolidation and workload optimization
  • Optimized end-to-end inference latency from ~10s to ~4s (60% reduction) by tuning model execution paths with TensorRT and ONNX Runtime
  • Designed and maintained 100+ production microservices using FastAPI, deployed with Docker on Kubernetes via Terraform-managed infrastructure with automated CI/CD
  • Implemented production observability using Elastic Stack and Grafana dashboards — request tracing, structured logging, performance metrics, and alerting
  • Deployed and tuned models via NVIDIA Triton Inference Server and vLLM for production-grade latency and throughput across vision and LLM workloads
YOLO Mask R-CNN PaddleOCR ONNX Runtime vLLM TensorRT NVIDIA Triton Gemini GPT Claude LLaMA Qwen Python FastAPI Cython Kubernetes Docker Terraform CI/CD Elasticsearch Grafana MongoDB PostgreSQL

Python / Machine Learning Engineer

Machine Learning 1
Jan 2022 — Oct 2022 Lahore, Pakistan Full-time

Developed and deployed OCR-based document processing pipelines with extensive preprocessing and post-processing logic to improve accuracy and robustness.

  • Built ML-backed REST APIs using Flask and FastAPI to serve real-time inference workloads
  • Automated model training workflows, including dataset preparation, annotation coordination, and validation processes
  • Improved OCR accuracy, throughput, and system reliability across multiple document types used in production
  • Created microservices for ML models with independent problem-solving capabilities
  • Collected, cleaned, and annotated data using VIA annotator for model training
  • Implemented requirements and data analysis to support decision-making processes
Python Flask FastAPI Machine Learning OCR Tesseract PaddleOCR OpenCV Pandas Data Processing

Data Science Trainee

Programmers Force
Sep 2021 — Dec 2021 Lahore, Pakistan Internship

Intensive training program focused on Python development, data science fundamentals, and object detection models.

  • Wrote effective and scalable Python code
  • Collected, organized, and interpreted data for actionable insights
  • Studied and researched Object Detection models (MRCNN, YOLO)
  • Created Flask endpoints to run Object Detection models
  • Built dummy services and ML pipelines
  • Integrated pipelines into Django projects
Python Flask Django YOLO MRCNN Data Analysis Object Detection
04

Technical Skills

MLOps & Platform

Kubernetes (60+ clusters) Docker CI/CD (Rolling & Canary) GPU-aware Scheduling Model Registry & Versioning Reproducible ML Pipelines

GPU & Inference Optimization

vLLM (PagedAttention) TensorRT NVIDIA Triton ONNX Runtime Model Quantization (FP16, 4-bit, LoRA) VRAM Optimization

AI / Machine Learning

YOLO Mask R-CNN Detectron2 DensePose Vision-Language Models LLMs (GPT, Gemini, Claude, Qwen, LLaMA)

Computer Vision & OCR

OpenCV PaddleOCR Tesseract DensePose NudeNet Image Classification

Backend & Data

Python FastAPI (REST/gRPC) Cython Elasticsearch MongoDB PostgreSQL RabbitMQ

Cloud & Infrastructure as Code

AWS (Lambda, EKS) Azure Functions Terraform Hybrid Cloud Serverless On-premise Deployments

Observability & Monitoring

Elastic Stack (ELK) Grafana Prometheus Request Tracing Alerting Cost-per-inference Tracking

Data Engineering

ETL Pipelines Pandas Web Scraping Selenium Event-driven Architectures Async Pipelines

Security & Compliance

Encrypted Data Pipelines CIS Hardening Audit Logging Compliance (KYC) Production Incident Management
05

Key Achievements

Milestones and recognitions throughout my career

Built SHEIN Product Scraper on Apify Platform

Developed production-grade web scraping solution with 95%+ success rate, generating $300-500/month recurring revenue from multiple clients worldwide

  • Earning $300-500/month in passive recurring revenue
  • 95%+ success rate with 3-layer anti-blocking system
  • Sophisticated captcha bypass with automatic IP rotation
  • Serves 150M+ SHEIN active user data for market research
  • 10-30s per product extraction with smart recovery

Top Rated Plus Freelancer

Achieved Top Rated Plus status on Upwork with 100% job success score across 50+ projects

  • 100% Job Success Score maintained over 4 years
  • Top Rated Plus badge (top 3% of all freelancers)
  • 50+ successfully completed projects
  • 5.0 average client rating across all projects
  • Zero disputes or negative feedback

4+ Years in MLOps

Extensive experience building production ML systems for enterprise clients across various industries

  • Built CI/CD pipelines for ML model deployment
  • Expertise in Docker, Kubernetes, and AWS/Azure
  • Automated model training and monitoring systems
  • Reduced deployment time from weeks to hours
  • Implemented MLflow, Airflow, and Kubeflow

15+ Happy Clients

Successfully delivered AI/ML solutions to clients worldwide, maintaining long-term partnerships

  • Clients across USA, UK, Europe, and Middle East
  • 80% repeat client rate for additional projects
  • Average project value increased by 200%
  • Long-term partnerships spanning 2+ years
  • Multiple 5-star reviews and referrals

Production AI Systems

Deployed 20+ ML models serving millions of predictions with 99.9% uptime

  • 20+ ML models deployed in production environments
  • Processing 10M+ predictions per month
  • 99.9% uptime with robust monitoring systems
  • Computer vision, NLP, and recommendation systems
  • Optimized inference time by 70% on average
06

Reviews & Testimonials

What clients and colleagues say about working with me

Mustapha Yakubu

CEO at Careerli

Upwork

Working with Noor was pleasant. We had multiple calls and he was gracious enough to go out of his way to make suggestions for what could be improved. He was also very patient when we experienced delays on our end. He is technically strong and always open to constant communication throughout the project. I am planning to work with him to complete other tasks.

Project: AI Resume & Cover Letter Platform

David Wanis

Client

Upwork

Noor was excellent. Kept going until the project worked as per specification.

Project: Azure Function ETL with AI/LLM API

Muhammad Shahid Shah

Senior Mobile App Developer | Technical Lead

Linkedin

Great experience working with Noor. He demonstrated strong technical skills and delivered quality results on our AI driven data scraping project.

Project: Shein AI Modesty Classification System

Waleed Iqbal Butt

Business Development & Digital Marketing Associate

Linkedin

Happy to recommend Noor. During our work together, he supported me with clear, reliable MLOps solutions that made our projects run smoother. As someone in digital marketing and business development, I appreciated how simply he explained technical steps and how quickly he helped us move forward. He's dependable, skilled, and great to work with.

Project: Cross-functional MLOps Collaboration

Mustapha Yakubu

CEO at Careerli

Upwork

Working with Noor was pleasant. We had multiple calls and he was gracious enough to go out of his way to make suggestions for what could be improved. He was also very patient when we experienced delays on our end. He is technically strong and always open to constant communication throughout the project. I am planning to work with him to complete other tasks.

Project: AI Resume & Cover Letter Platform

David Wanis

Client

Upwork

Noor was excellent. Kept going until the project worked as per specification.

Project: Azure Function ETL with AI/LLM API

Muhammad Shahid Shah

Senior Mobile App Developer | Technical Lead

Linkedin

Great experience working with Noor. He demonstrated strong technical skills and delivered quality results on our AI driven data scraping project.

Project: Shein AI Modesty Classification System

Waleed Iqbal Butt

Business Development & Digital Marketing Associate

Linkedin

Happy to recommend Noor. During our work together, he supported me with clear, reliable MLOps solutions that made our projects run smoother. As someone in digital marketing and business development, I appreciated how simply he explained technical steps and how quickly he helped us move forward. He's dependable, skilled, and great to work with.

Project: Cross-functional MLOps Collaboration
07

Services I Offer

End-to-end solutions for your AI and infrastructure needs

MLOps & AI Solutions

End-to-end ML pipeline development, model deployment, and infrastructure automation for production-ready AI systems.

  • ML Pipeline Design
  • Model Training & Fine-tuning
  • Production Deployment

Data Engineering

Build robust ETL pipelines, data warehouses, and real-time processing systems for actionable insights.

  • ETL Pipeline Development
  • Database Architecture
  • Data Processing at Scale

Computer Vision

Custom object detection, image classification, OCR systems, and visual AI solutions for your business.

  • Object Detection (YOLO)
  • OCR & Document Processing
  • Image Classification

DevOps & Automation

CI/CD pipelines, containerization, infrastructure automation, and cloud deployment solutions.

  • CI/CD Implementation
  • Docker & Kubernetes
  • Cloud Infrastructure

Web Scraping & Automation

Large-scale data extraction, browser automation, and anti-detection scraping solutions.

  • Anti-Bot Bypass
  • Proxy Management
  • Data Extraction Pipelines

API Development

RESTful APIs, microservices architecture, and backend systems using FastAPI and Flask.

  • FastAPI / Flask APIs
  • Microservices Design
  • WebSocket Integration
08

Get In Touch

Have a project in mind? Let's create something amazing together.

Email Me

Nooraftab779@gmail.com

WhatsApp

+92 331 4816207

LinkedIn

Connect with me

Upwork

Hire me

Book Meeting

Schedule a call

Or find me on